The DOI matches its Crossref record; 27 of the 31 references it cites were found in a registry or an index, and 4 are not checked yet; 2 of the 6 sources its claims cite were checked for support.
Safe Reinforcement Learning via Shielding
Open accessAbstract
Reinforcement learning algorithms discover policies that maximize reward, but do not necessarily guarantee safety during learning or execution phases. We introduce a new approach to learn optimal policies while enforcing properties expressed in temporal logic. To this end, given the temporal logic specification that is to be obeyed by the learning system, we propose to synthesize a reactive system called a shield. The shield monitors the actions from the learner and corrects them only if the chosen action causes a violation of the specification. We discuss which requirements a shield must meet to preserve the convergence guarantees of the learner. Finally, we demonstrate the versatility of our approach on several challenging reinforcement learning scenarios.
Verified 2026-10-24 · 4 references not checked yet
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
The paper proposes shielded reinforcement learning, where a reactive shield synthesized from a temporal-logic safety specification and an environment abstraction monitors learner actions and replaces them only when needed to avoid unsafe behavior. It proves correctness and minimal interference, gives conditions for preserving convergence, and demonstrates safety and often faster learning in four simulated domains.
Key findings
- Shields can enforce safety during both learning and execution without changing the reward signal, and they are minimally interfering: they block an action only when it could lead to a specification violation under the abstraction.
- In the water-tank example, the product game had 602 states, and the shield prevented opening the inflow valve above level 93 and forced opening when the level became too low.
- Shield synthesis was fast in the reported experiments: under 2 seconds for the grid-world cases, about 2 seconds for the self-driving car, under 1 second for the water tank, and under 7 seconds for Pacman.
- In the 15x9 grid world, only the shielded setting with a ranking of 3 actions and no penalty found the optimal path; in the water-tank experiments, shielded Q-learning and SARSA reached optimal policies in significantly fewer episodes than unshielded versions; in the self-driving car case, the shielded learner never crashed and learned faster; in Pacman, the shielded agent cleared the game from the first episode.
Methods
- The approach specifies safety in linear temporal logic, converts it to a deterministic safety automaton, builds a conservative finite-state abstraction of the MDP environment, and solves a safety game to synthesize a shield.
- The shield receives the learner’s action or ranked list of actions and executes the highest-ranked safe action; it deviates only if every ranked action is unsafe, in which case it chooses some safe action.
- The paper proves correctness and minimal interference, constructs a product MDP to argue convergence for standard MDP learners, and evaluates the method in simulations: 9x9 and 15x9 grid worlds, a 480x480-pixel self-driving car task, the water-tank controller, and an 18x5 Pacman-like maze, using tabular Q-learning, deep Q-networks, and approximate Q-learning.
Limitations
- The paper states that the main downside is the need for an approximate model or abstraction indicating when actions are unsafe, and argues this is unavoidable when allowed actions depend on the environment state.
- The convergence argument requires that when the learner’s ranking contains no safe action, the shield’s choice among safe actions follows a fixed probability distribution over time, and it assumes the learner can observe the shield state.
- Evident from the text: the evaluation is entirely simulation-based, with no real-world deployment; the approach relies on the abstraction being conservative, and the shield must remain active after learning because unsafe actions may still be part of the learned policy.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
45 Own interpretation · 10 Own result · 4 No basis given · 1 Partly supported · 1 Not supported · 1 No full text of the sources · 1 Not checked yet · accuracy not yet measured
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The results show that only the unshielded versions experience negative rewards.”
§ Safe Reinforcement Learning via Shielding > Experiments
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Fig. 11 shows that both shielded (dashed lines) and unshielded Q-learning and SARSA experiments (solid lines) do reach an optimal policy.”
§ Safe Reinforcement Learning via Shielding > Experiments
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In all of them, the learning performance of the shielded agents improved compared to the unshielded case.”
§ Safe Reinforcement Learning via Shielding > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As expected, in the shielded case, pacman is always able to clear the game starting from the first episode.”
§ Safe Reinforcement Learning via Shielding > Experiments
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, the shielded implementations reach the optimal policy in a significantly shorter time than the unshielded implementations.”
§ Safe Reinforcement Learning via Shielding > Experiments
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the computational cost of the construction of the shield depends on the complexity of the specification and a very abstract version of the system”
§ Safe Reinforcement Learning via Shielding > Safety in Formal Methods.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“since the shield corrects unsafe actions in the learning and deployment phases.”
§ Safe Reinforcement Learning via Shielding > Related Work Safety in RL.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Even though the inner working of learning algorithms is often complex, shielding with respect to critical safety specifications may be manageable (as we demonstrate in upcoming sections).”
§ Safe Reinforcement Learning via Shielding > Environment
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, to reason about when exactly a specification violation cannot be avoided, we have to give a (coarse finite-state) abstraction of the environment dynamics.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We introduce a new approach to learn optimal policies while enforcing properties expressed in temporal logic.”
§ Safe Reinforcement Learning via Shielding > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Probably not a substantial claim. 17
- It may state two separate claims as one. 17
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“An exploration process is called safe if no undesirable states are ever visited, which can only be achieved through the incorporation of external knowledge (García and Fernández 2015;Moldovan and Abbeel 2012).”
§ Safe Reinforcement Learning via Shielding > Related Work Safety in RL.
What the sources say
- Not checked — no full text of the sourcecites García and Fernández 2015
- Disputedcites Moldovan and Abbeel 2012 sourceExperimental — accuracy not yet measured
The Safety Constraint section (p. 3) defines safety as preserving ergodicity with a user-specified probability, not as never visiting an undesirable state. The Safety Counter-Examples section (p. 3) explicitly argues against defining safety by safe and unsafe states. The paper uses a prior belief and information learned during exploration, but does not say that incorporating external knowledge is the only way to achieve safety. (LLM, conf=0.82)
Checked by a reasoning model — may be wrong.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Unlike the approach in the paper, their technique does not work with function approximation, which prevents it from being used in complex scenarios.”
§ Safe Reinforcement Learning via Shielding > Safety in Formal Methods.
What the sources say
- Partly supportedcites Wen, Ehlers, and Topcu 2015 sourceExperimental — accuracy not yet measured
Section V (Examples; page unavailable) shows that learning time and iterations grow as the grid grows, and describes a trade-off between performance and computation cost. This supports a scalability concern, but the paper does not say that its technique cannot work with function approximation or that it cannot be used in complex scenarios. (LLM, conf=0.76)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The plot in Fig. 10 shows that the accumulated rewards for unshielded RL (red, dashed) increases over time, but the car still crashes at the end of the simulation.”
§ Safe Reinforcement Learning via Shielding > Experiments
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In scenarios with |rank t | = 1 (red) or with penalties (solid), the agent computes a suboptimal path.”
§ Safe Reinforcement Learning via Shielding > Experiments
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Only the shielded version with |rank t | = 3 without penalty (blue, dashed) finds the optimal path, resulting in a higher average reward.”
§ Safe Reinforcement Learning via Shielding > Experiments
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Furthermore, the shielded version is not only safe, but also tends to learn more rapidly.”
§ Safe Reinforcement Learning via Shielding > Experiments
hedgedconditionalSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The shielded version without punishment (blue, solid) learns more rapidly than the unshielded learner and never crashes.”
§ Safe Reinforcement Learning via Shielding > Experiments
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“For specifications in LTL, it is known how to check if it is a safety language and how to compute a safety automaton that represents it (Kupferman and Vardi 2001).”
§ Safe Reinforcement Learning via Shielding > The set of all distributions on X is denoted by Distr(X).
What the sources say
- Not checked — no full text of the sourcecites Kupferman and Vardi 2001 source
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“An abstraction of an MDP describes how its executions can possibly evolve, and provides the needed information about the environment to allow planning ahead w.r.t. the safety properties of interest.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“and is independent of the state space components of the system to be controlled that are irrelevant for enforcing the safety specification.”
§ Safe Reinforcement Learning via Shielding > Safety in Formal Methods.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It ensures safety and minimum interference.”
§ Safe Reinforcement Learning via Shielding > Environment
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The shield is computed upfront from the safety part of the given system specification and an abstraction of the agent's environment dynamics.”
§ Safe Reinforcement Learning via Shielding > Environment
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This boundary helps to separate the concerns, e.g., safety and correctness on one side and convergence and optimality on the other”
§ Safe Reinforcement Learning via Shielding > The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The latter is necessary as the valve has to stay open for at least three time steps.So as the inflow may be up to 2 liters/second during this time and the outflow may be 0, there is otherwise an overflow risk.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This approximation may have finitely many states even if the MDP has infinitely many states and/or is only approximately known.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Last but not least, the shielding framework is compatible with mechanisms such as function approximation, employed by learning algorithms in order to improve their scalability.”
§ Safe Reinforcement Learning via Shielding > The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The algorithms we present for the computation of shields make relatively mild assumptions on the input-output structure of the learning algorithm (rather than its inner working).”
§ Safe Reinforcement Learning via Shielding > Environment
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Reactive synthesis enforces ϕ s by solving a safety game built from ϕ s and an abstraction of the environment in which the policy is to be executed.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Planning ahead does not require the environment dynamics to be completely known in advance.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“All of the above approaches have in common that a faithful, yet precise enough, abstraction of the physical environment is required, which is not only difficult to obtain in practice, but also introduces the mentioned computa-tional burden.”
§ Safe Reinforcement Learning via Shielding > Safety in Formal Methods.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“and provides a basis for the convergence analysis of a shielded RL algorithm.”
§ Safe Reinforcement Learning via Shielding > The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This ensures that paths that model traces that cannot occur in M are rejected by ϕ M as early as possible.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As the shield is generated using the specification, it plans ahead for this not to happen, so it must prevent the opening of the inflow valve if the level is above 93.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our setup introduces a clear boundary between the learning agent and the shield.”
§ Safe Reinforcement Learning via Shielding > The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Given that the environment is often represented as an MDP, such an abstraction has to be conservative w.r.t. the behavior of the real MDP.”
§ Safe Reinforcement Learning via Shielding > Safety, Abstractions, and Games
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
- Rests on an assumption: The environment is often represented as an MDP.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Control methods based on RL partly address this problem, but do not typically incorporate any correctness guarantees.”
§ Safe Reinforcement Learning via Shielding > Safety in Formal Methods.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Shielding exploits this possibility.”
§ Safe Reinforcement Learning via Shielding > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Since an unsafe action is always mapped to a safe one, this does not pose a problem and the agent never has to learn to avoid unsafe actions.Consequently, the shield is (again) needed during the learning and execution phases.”
§ Safe Reinforcement Learning via Shielding > Shielded Reinforcement Learning
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our experiments show, however, that in applications in which safe learning is needed, the effort to construct an abstraction is well-spent, as our approach not only makes learning safe, but also shows great promise of improving learning performance.”
§ Safe Reinforcement Learning via Shielding > Conclusion
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The agent assigns a punishment r t+1 < 0 to a 1 t and learns that selecting a 1 t at state s t is unsafe, without ever violating ϕ s .”
§ Safe Reinforcement Learning via Shielding > Shielded Reinforcement Learning
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- The authors state this as an assumption: The environment can be modeled as an MDP M = (S, s_I, A, P, R).
- The authors state this as an assumption: We have constructed an abstraction ϕ_M.
- The authors state this as an assumption: The learner accepts elements from S × Q as state input (for the state space of the shield Q).
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, there is no guarantee that unsafe actions are not part of the final policy.Therefore, the shield has to remain active even after the learning phase.”
§ Safe Reinforcement Learning via Shielding > Shielded Reinforcement Learning
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The authors state this as an assumption.
The specification ϕ is a set of properties {ϕ1,...,ϕl} such that L(ϕ) is the intersection of the languages of its properties.
“We assume ϕ is a set of properties {ϕ 1 , . . ., ϕ l } such that L(ϕ) = i L(ϕ i ).”
The paper relies on this without support.
The environment is often represented as an MDP.
“Given that the environment is often represented as an MDP, such an abstraction has to be conservative w.r.t. the behavior of the real MDP.”
The paper relies on this without support.
Safety or more generally correctness during learning and execution of controllers has attracted significantly less attention than convergence, optimality and data-efficiency of learning algorithms.
“While convergence, optimality and data-efficiency of learning algorithms are relatively well understood, safety or more generally correctness during learning and execution of controllers has attracted significantly less attention.”
The authors state this as an assumption.
The environment can be modeled as an MDP M = (S, s_I, A, P, R).
“Assumption 1 (i) The environment can be modeled as an MDP M = (S, s I , A, P, R).”
The authors state this as an assumption.
We have constructed an abstraction ϕ_M.
“(ii) We have constructed an abstraction ϕ M .”
The authors state this as an assumption.
The learner accepts elements from S × Q as state input (for the state space of the shield Q).
“(iii) The learner accepts elements from S × Q as state input (for the state space of the shield Q).”
The authors state this as an assumption.
Some abstraction ϕ_M = (Q_M, q_0,M, A × L, δ_M, F_M) of M for some MDP observer function f : S → L is given.
“We assume some abstraction ϕ M = (Q M , q 0,M , A × L, δ M , F M ) of M for some MDP observer function f : S → L to be given.”
The authors state this as an assumption.
ϕ_s has Σ = L × A, i.e., it reads the part of the system behavior that the abstraction is concerned with.
“Since ϕ s models a restriction of the traces of the MDP and the learner together that we want to enforce, we assume it to have Σ = L × A, i.e., it reads the part of the system behavior that the abstraction is concerned with.”
The authors state this as an assumption.
A shield S = (Q_S, q_0,S, Σ_I,S, Σ_O,S, δ_S, λ_S) is given for M and for some MDP labeling function f : S → L.
“Assume also that a shield S = (Q S , q 0,S , Σ I,S , Σ O,S , δ S , λ S ) is given for M and for some MDP labeling function f : S → L.”
The authors state this as an assumption.
Whenever an action ranking is chosen by the learner that does not contain a safe action, there is a fixed probability distribution over the safe actions executed instead, constant over time.
“Note that this argument requires that whenever an action ranking is chosen by the learner that does not contain a safe action, there is a fixed probability distribution over the safe actions executed instead.”
The authors state this as an assumption.
The learner has access to the state of the shield and can base its actions on it.
“Note that we made use of the fact that the learner has access to the state of the shield and can base its actions on it in this argument.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“To provide system-level, physically absolute safety guarantees, SafeCodeRL incorporates a hard shielding mechanism [37].”
From SafeCodeRL: Security-Constrained Multi-Agent Reinforcement Learning for Trustworthy LLM-Generated IoT/CPS Software (2026) · cites this paper as [37]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Safe Reinforcement Learning via Shielding > Synthesis of Shields
“ϕ M = (Q M , q 0,M , A × L, δ M , F M )”
§ Safe Reinforcement Learning via Shielding > Synthesis of Shields
“(δ(q, (l, a)), δ M (q, (l, a))), for all (q, q M ) ∈ G, l ∈ L, a ∈ A”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Safe Reinforcement Learning via Shielding > Synthesis of Shields
“if we merge all states in F × Q M into a single error state”
§ Safe Reinforcement Learning via Shielding > Synthesis of Shields
“F g = (F × Q M ) ∪ (Q × (Q M \ F M )).”
Citations
Cites
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. (2005) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2004.842673 · Cited by 25,268
(2008) · Cited by 4,811
Handbook of theoretical computer science
Leeuwen, J. V. (1990) · Medical Entomology and Zoology · Cited by 4,259
Valiant, L. G. (1984) · ACM eBooks · doi:10.1145/3749740.3749746 · Cited by 4,211
Valiant, L. G. (1984) · Communications of the ACM · doi:10.1145/1968.1972 · Cited by 3,279
Emerson, E. (1991) · Handbook of Theoretical Computer Science, Volume B: Formal Models and Sematics · doi:10.1016/b978-0-444-88074-1.50021-4 · Cited by 3,268
The Arcade Learning Environment: An Evaluation Platform for General Agents
Bellemare, M. G., Naddaf, Y., Veness, J., & Bowling, M. (2013) · Journal of Artificial Intelligence Research · doi:10.1613/jair.3912 · Cited by 2,161
The temporal logic of programs
Pnueli, A. (1977) · 18th Annual Symposium on Foundations of Computer Science (sfcs 1977) · doi:10.1109/sfcs.1977.32 · Cited by 1,523
Cited by
Sorted by citations
Items 501–600 of 700
SEO: Safety-Aware Energy Optimization Framework for Multi-Sensor Neural Controllers at the Edge
Odema, M., Ferlez, J., Shoukry, Y., et al. (2023) · Proceedings - ACM IEEE Design Automation Conference · doi:10.1109/dac56929.2023.10247751 · Cited by 1
Risk-Sensitive Inhibitory Control for Safe Reinforcement Learning
Lederer, A., Noorani, E., Baras, J. S., et al. (2023) · Proceedings of the IEEE Conference on Decision & Control, including the Symposium on Adaptive Processes · doi:10.1109/cdc49753.2023.10383524 · Cited by 1
Computationally Efficient Reinforcement Learning: Targeted Exploration leveraging Simple Rules
Natale, L. D., Svetozarevic, B., Heer, P., & Jones, C. N. (2023) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc49753.2023.10383956 · Cited by 1
Scalable Semantic Non-Markovian Simulation Proxy for Reinforcement Learning
Mukherji, K., Parkar, D., Pokala, L., Aditya, D., Shakarian, P., & Dorman, C. (2024) · IEEE International Conference on Semantic Computing (ICSC) · doi:10.1109/icsc59802.2024.00035 · Cited by 3
Safe Reinforcement Learning Through Regret and State Restorations in Evaluation Stages
Gros, T. P., Müller, N. J., Höller, D., & Wolf, V. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-75778-5_2 · Cited by 1
Maurer, F., Meidinger, M., Lu, Y., et al. (2024) · IEEE Nordic Circuits and Systems Conference (NorCAS) · doi:10.1109/norcas64408.2024.10752457 · Cited by 1
Investigating the Impact of Choice on Deep Reinforcement Learning for Space Controls
Hamilton, N. P., Dunlap, K. M., & Hobbs, K. L. (2024) · IEEE International Conference on Space Mission Challenges for Information Technology · doi:10.1109/smc-it61443.2024.00014 · Cited by 1
Repairing Neural Networks for Safety in Robotic Systems using Predictive Models
Majd, K., Clark, G., Fainekos, G., et al. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802650 · Cited by 1
Towards Blockchain-based AI Alignment Using a Proof of Personhood Consensus Mechanism
Neulinger, A., & Sparer, L. (2024) · International Conference on Blockchain Computing and Applications (BCCA) · doi:10.1109/bcca62388.2024.10844393 · Cited by 1
Society Protect: Developing A Voting Mechanism for Blockchain-based AI Alignment
Neulinger, A., Sparer, L., Roshanaei, M., & Ramljak, D. (2024) · International Conference on Blockchain Computing and Applications (BCCA) · doi:10.1109/bcca62388.2024.10844386 · Cited by 1
Hamilton, N. P., Dunlap, K. M., Viramontes, F. O., Landauer, D. C., Kain, E. T., & Hobbs, K. L. (2025) · AIAA SciTech Forum · doi:10.2514/6.2025-1144 · Cited by 1
Learning-Based Shielding for Safe Autonomy under Unknown Dynamics
Reed, R., & Lahijanian, M. (2025) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc63710.2025.11108000 · Cited by 4
A Human-in-the-Loop Approach to Learning Social Norms and Behavioural Policies
Deane, O., & Ray, O. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-08333-3_3 · Cited by 1
Topological Approximate Dynamic Programming under Temporal Logic Constraints
Li, L., & Fu, J. (2019) · Proceedings of the IEEE Conference on Decision & Control, including the Symposium on Adaptive Processes · doi:10.1109/cdc40024.2019.9029959 · Cited by 1
Near-Optimal Reactive Synthesis Incorporating Runtime Information
Bharadwaj, S., Vinod, A. P., Dimitrova, R., et al. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196581 · Cited by 1
Self-citationTemporal-Logic-Based Reward Shaping for Continuing Reinforcement Learning Tasks
Jiang, Y., Bharadwaj, S., Wu, B., et al. (2021) · Texas Digital Library (University of Texas) · doi:10.26153/tsw/11579 · Cited by 1
Self-citationLifted model checking for relational MDPs
Yang, W.-C., Raskin, J.-F., & Raedt, L. D. (2022) · Machine Learning · doi:10.1007/s10994-021-06102-7 · Cited by 1
Lane Keeping Algorithm for Autonomous Driving via Safe Reinforcement Learning
Kong, Q., Zhang, L., & Xu, X. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-82153-1_36 · Cited by 1
Safe Learning for Near-Optimal Scheduling
Busatto-Gaston, D., Chakraborty, D., Guha, S., Pérez, G. A., & Raskin, J.-F. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-85172-9_13 · Cited by 1
Enforcing Hard State-Dependent Action Bounds on Deep Reinforcement Learning Policies
Cooman, B. D., Suykens, J. A. K., & Ortseifen, A. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-25891-6_16 · Cited by 1
Formal Verification of End-to-End Learning in Cyber-Physical Systems: Progress and ChallengesPreprint
Fulton, N., Hunt, N., Hoang, N., & Das, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.09181 · Cited by 5
Extended Radial Basis Function Controller for Reinforcement LearningPreprint
Capel, N., & Zhang, N. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2009.05866 · Cited by 1
Runtime Monitoring for Markov Decision ProcessesPreprint
Junges, S., Torfah, H., & Seshia, S. A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.12322 · Cited by 1
Zhang, X., Peng, Y., Luo, B., et al. (2023) · doi:10.36227/techrxiv.17205740.v3 · Cited by 1
CESAR: Control Envelope Synthesis via Angelic RefinementsPreprint
Kabra, A., Laurent, J., Mitsch, S., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2311.02833 · Cited by 1
The Autonomy-Alignment Problem in Open-Ended Learning Robots Formalising the Purpose FrameworkPreprint
Baldassarre, G., Duro, R. J., Cartoni, E., Khamassi, M., Romero, A. M., & Santucci, V. G. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5547070 · Cited by 1
Neuro-symbolic deep reinforcement learning for safe urban driving using low-cost sensors.
Albilani, M. (2024) · theses.fr (ABES) · doi:10.70675/ea6f5d11z7d78z4484zb809z38299a88535e · Cited by 1
Safe Learning for Near Optimal SchedulingPreprint
Geeraerts, G., Guha, S., Pérez, G. A., & Raskin, J.-F. (2020) · arXiv (Cornell University)
FISAR: Forward Invariant Safe Reinforcement Learning with a Deep Neural Network-Based OptimizePreprint
Sun, C., Kim, D.-K., & How, J. P. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.11419
Adversarial Attacks in a Deep Reinforcement Learning based Cluster Scheduler
Zhang, S.-J., Wang, C., & Zomaya, A. Y. (2020) · Proceedings - International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems · doi:10.1109/mascots50786.2020.9285955
What criminal and civil law tells us about Safe RL techniques to generate law-abiding behaviour.
Ashton, H. (2020) · UCL Discovery (University College London)
Self-Repairing Neural Networks: Provable Safety for Deep Networks via Dynamic Repair.Preprint
Leino, K. R. M., Fromherz, A., Mangal, R., Fredrikson, M., Parno, B. J., & Păsăreanu, C. S. (2021) · arXiv (Cornell University)
Neural Network Repair with Reachability AnalysisPreprint
Yang, X.-D., Yamaguchi, T., Tran, H.-D., Hoxha, B., Johnson, T. T., & Prokhorov, D. (2021) · International Conference on Formal Modeling and Analysis of Timed Systems · doi:10.48550/arxiv.2108.04214 · Cited by 38
Dependability Analysis of Deep Reinforcement Learning based Robotics and Autonomous Systems.Preprint
Dong, Y., Zhao, X., & Huang, X. (2021) · arXiv (Cornell University)
Uniformly Conservative Exploration in Reinforcement LearningPreprint
Xu, W., Ma, J. Y., Xu, K., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.13060
MoveRL: To a Safer Robotic Reinforcement Learning Environment
Liu, G., Winter, J. D., Vanderborght, B., et al. (2022) · Communications in computer and information science · doi:10.1007/978-3-030-93842-0_14
Zhang, X., Peng, Y., Luo, B., et al. (2021) · doi:10.36227/techrxiv.17205740.v1
Zhang, X., Peng, Y., Luo, B., et al. (2021) · doi:10.36227/techrxiv.17205740
Dependable Reinforcement Learning via Timed Differential Dynamic Logic
Wang, R., Zhang, Y., Sun, H., et al. (2021) · Proceedings - IEEE Symposium on Computers and Communications/IEEE Symposium on Computers and Communications · doi:10.1109/iscc53001.2021.9631442
Refinement Of Reinforcement Learning Algorithms Guided By Counterexamples
Gangopadhyay, B., Vishnoi, S., & Dasgupta, P. (2022) · IEEE Women in Technology Conference (WINTECHCON) · doi:10.1109/wintechcon55229.2022.9832063
A Safe Training Approach for Deep Reinforcement Learning-based Traffic Engineering
Wang, L., Wang, M., & Zhang, Y. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838944
Task Independent Safety Assessment for Reinforcement Learning
Jocas, M., Zoghlami, F., Kurrek, P., et al. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-15908-4_16
Leveraging Non-Experts and Formal Methods to Automatically Correct Robot Failures
van Waveren, S. (2022) · ACM/IEEE International Conference on Human-Robot Interaction (HRI) · doi:10.1109/hri53351.2022.9889361
Intelligent and Dependable Decision-Making Under Uncertainty
Jansen, N. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-27481-7_3
Safe Online Integral Reinforcement Learning for Control Systems via Controller Decomposition
Sun, J., Song, X., & Ling, R. (2023) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-023-08026-x
Zhang, X., Peng, Y., Luo, B., et al. (2023) · doi:10.36227/techrxiv.17205740.v4
An Aeroengine Control Method Based on Safe Reinforcement Learning
Yu, Z., Huang, Z., Yan, S., et al. (2023) · International Conference on Control Science and Systems Engineering (ICCSSE) · doi:10.1109/iccsse59359.2023.10245899
den Hengst, F. (2023) · doi:10.5463/thesis.390
Safe Integration of Learning in SystemC using Timed Contracts and Model Checking
Blohm, P., Adelt, J., & Herber, P. (2023) · ACM/IEEE International Conference on Formal Methods and Models for System Design (MEMOCODE) · doi:10.1145/3610579.3611078
Decomposing user-defined tasks in a reinforcement learning setup using TextWorld
Petsanis, T., Keroglou, C., Kapoutsis, A. C., Kosmatopoulos, E. B., & Sirakoulis, G. C. (2023) · Frontiers in Robotics and AI · doi:10.3389/frobt.2023.1280578
Zhong, B.-Z., Liu, S.-Y., Caccamo, M., & Zamani, M. (2023) · IEEE Conference on Decision and Control · doi:10.1109/cdc49753.2023.10384154 · Cited by 5
Jin, C., Ma, X., Ren, T., Lin, W., & Ding, Z. (2024) · ACM International Conference on Hybrid Systems: Computation and Control (HSCC) · doi:10.1145/3641513.3650137
Towards human-level artificial intelligence agents
Leung, J. C. (2024) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/174532
Student Research Abstract: Enhancing Safety in Cyber-Physical Systems Through Runtime Enforcement
Lopez-Miguel, I. D. (2024) · ACM/SIGAPP Symposium on Applied Computing (SAC) · doi:10.1145/3605098.3635171
Formal Control Synthesis via Safe Reinforcement Learning Under Real-Time Specifications
Lv, P., Luo, G., He, Z., et al. (2024) · doi:10.1109/icca62789.2024.10591902
Towards Enhancing Driver’s Perceived Safety in Autonomous Driving: A Shield-Based Approach
Abe, R., Cai, J., Wang, T., et al. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-64850-2_15
Xu, W., Zhao, Y., & Tan, H. (2024) · International Symposium on Intelligent Unmanned Systems and Artificial Intelligence (SIUSAI) · doi:10.1145/3669721.3669735
Czarnecki, K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-68738-9_32
Approaching Single-Episode Survival Reinforcement Learning with Safety-Threshold Q-Learning
Perotto, F. S., Nargeot, M., & Ouahbi, A. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-77941-1_12
An Adaptive Interpretable Safe-RL Approach for Addressing Smart Grid Supply-Side Uncertainties
Dey, S., Verma, P., Dasgupta, P., & Dey, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-70074-3_7
Towards Verification-Driven Control Learning
Mitsch, S. (2024) · International Conference on Applied Algorithms · doi:10.1109/icaa64256.2024.00023
Data-Driven Safe Control of Stochastic Nonlinear Systems
Esmaeili, B., & Modares, H. (2024) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2025.01.021
Enhance Exploration in Safe Reinforcement Learning with Contrastive Representation Learning
Doan, D. K., Le, B. G., & Ta, V. C. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-5887-9_1
Fujita, S. T., Saruchi, S. A., Al-Talib, A. A. M., Wahid, N., Chowdhury, A. K., & Maidin, S. N. (2025) · Proceedings of International Conference on Artificial Life and Robotics · doi:10.5954/icarob.2025.os13-5
Enhancing Safety in Autonomous Racing With Constrained Reinforcement Learning
Yu, K., Fu, M., Zhang, T., et al. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3566591
Safe Switching Model-Free Value Iteration for General Nonlinear Systems
Lala, T. (2025) · IEEE Access · doi:10.1109/access.2025.3567496
Safe, Optimal and Compact Strategies for Cyber Physical Systems
Taankvist, J. H. (2025) · doi:10.54337/aau779529764
MEDRAIL: MEDical Reinforcement-based Agentic Intelligence for Lifesciences (Preprint)Preprint
Sharma, P. (2025) · doi:10.2196/preprints.78741
Walker, T. T., & Ide, J. S. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-8889-0_15
Polychronous RSS in a Process-Algebraic Framework - A Case Study
Crisafulli, P., Durier, A., Puyobro, B., et al. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-94533-5_15
Completing Structured Arguments in Assumption-Based Argumentation
Popescu, A., & Wallner, J. P. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-04587-4_7
Reusable Test Suites for Reinforcement Learning
Betten, J. E., Mazouni, Q., Gross, D., Lind, P. G., & Spieker, H. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-05188-2_9 · Cited by 1
Compositional Shield Synthesis for Safe Reinforcement Learning in Partial Observability
Carr, S., Bakirtzis, G., & Topcu, U. (2025) · IEEE Open Journal of Control Systems · doi:10.1109/ojcsys.2025.3611725
Self-citationA Parametric Model for Near-Optimal Online Synthesis with Robust Reach-Avoid Guarantees
Gleirscher, M., & Hönnecke, P. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-01377-4_14
Scenario-Based Hierarchical Reinforcement Learning for Automated Driving Decision Making
Abdelhamid, M. Y., Vater, L., & Ajanović, Z. (2025) · doi:10.1109/icat66432.2025.11189264
Predictive Safety Shield for Dyna-Q Reinforcement Learning
Jin, P., Krasowski, H., & Vanneaux, E. (2025) · European Control Conference (ECC) · doi:10.23919/ecc65951.2025.11186958
Adversarial and Backdoor Threats in Autonomous-Vehicle and Embodied-AI Systems
Shafique, M. S., Hanif, M. A., Guesmi, A., et al. (2025) · IEEE International System-on-Chip Conference (SOCC) · doi:10.1109/socc66126.2025.11235443
Wete, E., Greenyer, J., Yaacov, T., et al. (2025) · ACM/IEEE International Conference on Model Driven Engineering Languages and Systems (MODELS) · doi:10.1109/models67397.2025.00010
Safe Multi-agent Reinforcement Learning Using Formal Runtime Enforcement: A Case Study
Mohapatra, V., Anand, A., & Pinisetty, S. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-11176-0_28
Reinforcement Learning for Optimizing FACTS Setpoints With Limited Set of Measurements
Tarle, M., Larsson, M., Ingeström, G., et al. (2025) · IEEE Open Access Journal of Power and Energy · doi:10.1109/oajpe.2025.3645591
Xie, H., Cao, H., & Zhao, H. (2025) · Systems · doi:10.3390/systems14010036
Safe Battery Use and Grid-Convenience in an Intelligent Energy Control System
Adelt, J., Herber, P., Hurink, J. L., Niehage, M., Remke, A., & Willemsen, L. (2026) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-032-06818-7_15
Rizehvandi, A., Azadi, S., & Eichberger, A. (2026) · World Electric Vehicle Journal · doi:10.3390/wevj17010033
Stochastic Reinforcement Learning with Stability Guarantees for Control of Unknown Nonlinear Systems
Quartz, T., Zhou, R., Sterck, H. D., et al. (2025) · Proceedings of the IEEE Conference on Decision & Control, including the Symposium on Adaptive Processes · doi:10.1109/cdc57313.2025.11312866
On the Challenges of Monitoring Robust Systems
Klampfl, L., Perko, A., & Wotawa, F. (2025) · International Conference on Dependable Systems and Their Applications (DSA) · doi:10.1109/dsa66321.2025.00031
Cost-Aware and Risk-Sensitive Learning for Autonomous Robots A Conceptual Framework
Eziokwu, U. J., Duruemeruo, U. C., Akande, N. A., et al. (2023) · International Journal of Scientific Research in Science Engineering and Technology · doi:10.32628/ijsrset23103204
Wang, Z., Li, R., & Niyato, D. T. (2026) · Complex & Intelligent Systems · doi:10.1007/s40747-026-02228-0
A Modular Reinforcement Learning Framework for Iterative FPS Agent Development
Lee, S. H., & Sung, H. (2026) · Electronics · doi:10.3390/electronics15030519
Clarifying Constraints in Interactive Robot Learning with Language Feedback
Kuehn, H., Santos, L., & Leite, I. (2026) · ACM/IEEE International Conference on Human-Robot Interaction (HRI) · doi:10.1145/3757279.3785583
Robust Adversarial Training for Sequential Decision Making in Safety-Critical Cyber-Physical SystemsPreprint
Amiri, S. M. H. (2026) · Research Square · doi:10.21203/rs.3.rs-9216120/v1
Ravi, R. V., K, P. K., & D, A. M. (2025) · doi:10.1109/icaft66710.2025.11453104
Robust Adversarial Training for Sequential Decision Making in Safety-Critical Cyber-Physical SystemsPreprint
Amiri, S. M. H. (2026) · doi:10.24108/preprints-3114775
Son, J., Park, M., Jeong, S. C., & Kim, S.-H. (2026) · Transactions of Korean Society of Automotive Engineers · doi:10.7467/ksae.2026.34.4.497
Oracle-Guided Soft Shielding for Safe Move Prediction in Chess
Rajendran, P. T., Arnéz, F., Espinoza, H., Delaborde, A., & Mraidha, C. (2025) · International Conference on Machine Learning and Applications · doi:10.1109/icmla66185.2025.00130
Assuring Autonomy Using MBSE, Formal Methods, and the Best Agile/DevOps Practices
Heitmeyer, C., & Leonard, E. I. (2026) · Springer eBooks · doi:10.1007/978-3-031-88546-4_3
Metadata
- Authors
- Alshiekh, M., Bloem, R., Ehlers, R., Könighofer, B., Niekum, S., & Topcu, U.
- Affiliations
- The University of Texas at Austin (US)
- Graz University of Technology (AT)
- University of Bremen (DE)
- Venue
- Proceedings of the AAAI Conference on Artificial Intelligence
- Published
- 2018-04-29
- Volume
- 32
- Issue
- 1
- ISSN
- 2159-5399 (ISSN-L)
- 2374-3468
- OpenAlex
- W2751422670
- Cited by
- 37
- References
- 19
- Citations by year
- 2026: 1 · 2025: 1 · 2024: 3 · 2023: 7 · 2022: 3 · 2021: 7 · 2020: 7 · 2019: 4 · 2018: 4
- FWCI
- 2.24
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Science Foundation (1652113, 1617639, 1646522)
- Austrian Science Fund (S11406-N23, W1255)
- Universität Bremen
- Defense Advanced Research Projects Agency (W911NF-16-1-0001, FA8650-15C)
- Army Research Office (W911NF-15-1-0592, W911NF-15-1-, W911NF-16-1-, W911NF, W911NF-15)
- Air Force Research Laboratory (FA8650-15-C-2546, FA8650-15-, 8650-16-C-2610)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Formal Methods in Verification
- Physical Sciences › Computer Science › Artificial Intelligence › Logic, Reasoning, and Knowledge
- Keywords
- safe reinforcement learning, temporal logic specifications, reactive systems, convergence guarantees, policy learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-24, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-24 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
