Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 76 of the 78 references it cites were found in a registry or an index.

Widening Access to Applied Machine Learning With TinyML

Open access

Abstract

Broadening access to both computational and educational resources is critical to diffusing machine learning (ML) innovation. However, today, most ML resources and experts are siloed in a few countries and organizations. In this article, we describe our pedagogical approach to increasing access to applied ML through a massive open online course (MOOC) on Tiny Machine Learning (TinyML). We suggest that TinyML, applied ML on resource-constrained embedded devices, is an attractive means to widen access because TinyML leverages low-cost and globally accessible hardware and encourages the development of complete, self-contained applications, from data collection to deployment. To this end, a collaboration between academia and industry produced a four part MOOC that provides application-oriented instruction on how to develop solutions using TinyML. The series is openly available on the edX MOOC platform, has no prerequisites beyond basic programming, and is designed for global learners from a variety of backgrounds. It introduces real-world applications, ML algorithms, data-set engineering, and the ethical considerations of these technologies through hands-on programming and deployment of TinyML applications in both the cloud and on their own microcontrollers. To facilitate continued learning, community building, and collaboration beyond the courses, we launched a standalone website, a forum, a chat, and an optional course-project competition. We also open-sourced the course materials, hoping they will inspire the next generation of ML practitioners and educators and further broaden access to cutting-edge ML technologies.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1162/99608f92.762d171a

Cites

76 of 76 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 We designed a new applied-ML course. The course is motivated by realworld applications, covering not only the software (algorithms) and hardware (embedded systems) but also the product life cycle and responsible AI considerations needed to deploy these applications. To make it globally accessible and scalable, we focused on the emerging TinyML domain and released the course as a MOOC on edX.
Figure 1 We designed a new applied-ML course. The course is motivated by realworld applications, covering not only the software (algorithms) and hardware (embedded systems) but also the product life cycle and responsible AI considerations needed to deploy these applications. To make it globally accessible and scalable, we focused on the emerging TinyML domain and released the course as a MOOC on edX.CC-BY · From the PDF
Figure 2 Tiny Machine Learning (TinyML) is focused on enabling new applications through the deployment of machine learning models onto embedded systems.
Figure 2 Tiny Machine Learning (TinyML) is focused on enabling new applications through the deployment of machine learning models onto embedded systems.CC-BY · From the PDF
Figure 3 Example TinyML devices: (a) Pico4ML, (b) Arduino Nano 33 BLE Sense, and (c) STMicroelectronics Sensor Tile. These and many other embedded devices are being built with integrated sensors, specifically for enabling on-device real-time data processing.
Figure 3 Example TinyML devices: (a) Pico4ML, (b) Arduino Nano 33 BLE Sense, and (c) STMicroelectronics Sensor Tile. These and many other embedded devices are being built with integrated sensors, specifically for enabling on-device real-time data processing.CC-BY · From the PDF
Figure 4 The number of courses is disproportionate to the number of systems.
Figure 4 The number of courses is disproportionate to the number of systems.CC-BY · From the PDF
Figure 5 The ML workflow from data collection to model training to inference. The spiral course design focuses on the neural-network model in Course 1, model application in Course 2, application deployment in Course 3, and, finally, Course 4 'closes the loop' by covering ML operations (MLOps) which enable the scaled management, deployment, and continual improvement of TinyML applications.
Figure 5 The ML workflow from data collection to model training to inference. The spiral course design focuses on the neural-network model in Course 1, model application in Course 2, application deployment in Course 3, and, finally, Course 4 'closes the loop' by covering ML operations (MLOps) which enable the scaled management, deployment, and continual improvement of TinyML applications.CC-BY · From the PDF
Figure 6 The Arudino TinyML Kit. Shown at left, the TinyML Kit includes the Arduino Nano 33 BLE Sense (Arduino, 2021a), an OV7675 camera module (ArduCam, 2021a), and a TinyML shield that simplifies sensor integration. Shown at right, the Nano 33 BLE includes a set of onboard integrated sensors (e.g., temperature/humidity, IMU, and microphone), a Bluetooth Low Energy (BLE) module, and an Arm Cortex-M microcontroller that can run neural-network models using TensorFlow Lite for Microcontrollers.
Figure 6 The Arudino TinyML Kit. Shown at left, the TinyML Kit includes the Arduino Nano 33 BLE Sense (Arduino, 2021a), an OV7675 camera module (ArduCam, 2021a), and a TinyML shield that simplifies sensor integration. Shown at right, the Nano 33 BLE includes a set of onboard integrated sensors (e.g., temperature/humidity, IMU, and microphone), a Bluetooth Low Energy (BLE) module, and an Arm Cortex-M microcontroller that can run neural-network models using TensorFlow Lite for Microcontrollers.CC-BY · From the PDF
Figure 7 Scaling TinyML using MLOps. We present a systematic approach for developing and deploying models into production. The MLOps stages are shown in a waterfall design, cascading down from one to another, which makes it easy to convey the important aspects within each stage. However, in practice, many organizations adopt an agile, iterative approach due to strong interdependencies that necessitate an agile flow.
Figure 7 Scaling TinyML using MLOps. We present a systematic approach for developing and deploying models into production. The MLOps stages are shown in a waterfall design, cascading down from one to another, which makes it easy to convey the important aspects within each stage. However, in practice, many organizations adopt an agile, iterative approach due to strong interdependencies that necessitate an agile flow.CC-BY · From the PDF
Image 8
CC-BY · From the PDF
Figure 9 Learner activities. We employ a wide array of learning activities to give students an immersive, self-paced online experience.
Figure 9 Learner activities. We employ a wide array of learning activities to give students an immersive, self-paced online experience.CC-BY · From the PDF
Figure 10 Students attack a pretrained KWS model. with malicious static noise. This triggers a spotting of the keyword "yes," revealing the need for security and privacy.
Figure 10 Students attack a pretrained KWS model. with malicious static noise. This triggers a spotting of the keyword "yes," revealing the need for security and privacy.CC-BY · From the PDF
Figure 11 Media Types. We used various video-production strategies throughout the course to maximize learning efficiency. (A) Picture-in-picture places a video clip in a small frame on top of another frame, playing them simultaneously. It enhances the perception of instructor presence while showing the student relevant content. (B) Split-screen, a slight variation, also improves the perception of instructor presence. (C) Tri-frames are useful for lab demos to enable 'hands-on' instruction from teaching assistants. (D) Screencasts coupled with associated lecture material reinforce concepts with code.
Figure 11 Media Types. We used various video-production strategies throughout the course to maximize learning efficiency. (A) Picture-in-picture places a video clip in a small frame on top of another frame, playing them simultaneously. It enhances the perception of instructor presence while showing the student relevant content. (B) Split-screen, a slight variation, also improves the perception of instructor presence. (C) Tri-frames are useful for lab demos to enable 'hands-on' instruction from teaching assistants. (D) Screencasts coupled with associated lecture material reinforce concepts with code.CC-BY · From the PDF
Figure 12 Course-enrollment metrics for Courses 1 through 3. Over 60,000 students are currently enrolled across all three courses. Course 4 has yet to open for enrollment. We expect another enrollment spike with 'Scaling TinyML.'
Figure 12 Course-enrollment metrics for Courses 1 through 3. Over 60,000 students are currently enrolled across all three courses. Course 4 has yet to open for enrollment. We expect another enrollment spike with 'Scaling TinyML.'CC-BY · From the PDF
Image 13
CC-BY · From the PDF
Figure 14 Education demographics based on voluntarily provided information. Many of our learners indicated an interest in TinyML to understand applied ML technologies to either pivot or grow further in their current positions.
Figure 14 Education demographics based on voluntarily provided information. Many of our learners indicated an interest in TinyML to understand applied ML technologies to either pivot or grow further in their current positions.CC-BY · From the PDF
Figure 15 Gender demographics based on voluntarily provided information. Collectively, we are working on engaging a more diverse population of learners with the aid of working groups that are part of the tinyML Foundation.
Figure 15 Gender demographics based on voluntarily provided information. Collectively, we are working on engaging a more diverse population of learners with the aid of working groups that are part of the tinyML Foundation.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 78 references

Research works printed but not linked: 9

  1. Bruner, J. S. (2009). The process of education. Harvard University Press.
  2. Chuvpilo, G. (2020). Who's ahead in AI research AT NeurIPS 2020? https://chuvpilo.medium. com/whos-ahead-in-ai-research-at-neurips-2020-bf2a40a54325
  3. Doyu, H., Morabito, R., & Höller, J. (2020). Bringing machine learning to the deepest IoT edge with TinyML as-a-service. IEEE IoT Newsl.
  4. Fosdick, A. (2021). Introduction to Embedded Systems Software and Development Environments. https://www.coursera.org/learn/introduction-embedded-systems Frame.io. (2021). Video review and collaboration software. Frame.io. https://www.frame.io/
  5. Grosz, B. J., Grant, D. G., Vredenburgh, K., Behrends, J., Hu, L., Simmons, A., & Waldo, J. (2019). Embedded ethiCS: Integrating ethics across CS curriculum [https: //cacm.acm. org/magazines/2019/8/238345-embedded-ethics/fulltext]. Communications of the ACM, 62 (8), 54-61.
  6. Jamieson, P. (2011). Arduino for teaching embedded systems. are computer scientists and engineer- ing educators missing the boat? Proceedings of the international conference on frontiers in education: computer science and computer engineering (FECS), 1.
  7. Maedche, A., Legner, C., Benlian, A., Berger, B., Gimpel, H., Hess, T., Hinz, O., Morana, S., & Söll- ner, M. (2019). AI-based digital assistants. Business & Information Systems Engineering, 61 (4), 535-544.
  8. Tiwari, V. (2010). MFCC and its applications in speaker recognition. International journal on emerging technologies, 1 (1).
  9. Zhang, G., Yan, C., Ji, X., Zhang, T., Zhang, T., & Xu, W. (2017). Dolphinattack: Inaudible voice commands. In Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security, 103-117.

Cited by

See all 77

Metadata

Authors
Janapa Reddi, V., Plancher, B., Kennedy, S., Moroney, L., Warden, P., Suzuki, L., Agarwal, A., Banbury, C., Banzi, M., Bennett, M., Brown, B., Chitlangia, S., Ghosal, R., Grafman, S., Jaeger, R., Krishnan, S., Lam, M., Leiker, D., Mann, C., Mazumder, M., Pajak, D., Ramaprasad, D., Smith, J. E., Stewart, M., & Tingley, D.
Affiliations
  • Edge Computing Lab, John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts, United States of America
  • Harvard University (US)
  • Google (United States) (US)
  • Arduino, Boston, Massachusetts, United States of America
  • Creative Class LLC, Westminster, Colorado, United States of America
Venue
Harvard Data Science Review
Publisher
MIT Press
Published
2022-01-27
Volume
4
Issue
1
ISSN
  • 2644-2353 (ISSN-L)
  • 2688-8513
OpenAlex
W3168641996
Cited by
62
References
78
Citations by year
2026: 6 · 2025: 8 · 2024: 11 · 2023: 30 · 2022: 6
FWCI
10.63
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Harvard Data Science Review (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • PhilPapers (PhilPapers Foundation) (submittedVersion)
  • arXiv (Cornell University)
Funders
Abdus Salam International Centre for Theoretical Physics
Topics
  • Physical Sciences › Computer Science › Computer Science Applications › Online Learning and Analytics
  • Physical Sciences › Computer Science › Computer Science Applications › Teaching and Learning Programming
  • Physical Sciences › Mathematics › Statistics and Probability › Statistics Education and Methodologies
Keywords
TinyML, massive open online courses, embedded devices, microcontroller, resource-constrained devices, applied machine learning, open educational resources, academia-industry collaboration, machine learning ethics, data engineering, community building

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.

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Widening Access to Applied Machine Learning With TinyML. https://app.cochat.ai/papers/10.1162/99608f92.762d171a
The paper
Janapa Reddi, V., Plancher, B., Kennedy, S., Moroney, L., Warden, P., Suzuki, L., Agarwal, A., Banbury, C., Banzi, M., Bennett, M., Brown, B., Chitlangia, S., Ghosal, R., Grafman, S., Jaeger, R., Krishnan, S., Lam, M., Leiker, D., Mann, C., . . . Tingley, D. (2022). Widening Access to Applied Machine Learning With TinyML. Harvard Data Science Review. https://doi.org/10.1162/99608f92.762d171a

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.