Adaptive multi-teacher multi-level knowledge distillation
Open accessThe DOI matches its Crossref record; 8 of the 8 references it cites were found in a registry or an index.
Abstract
Abstract Knowledge distillation (KD) is an effective learning paradigm for improving the performance of lightweight student networks by utilizing additional supervision knowledge distilled from teacher networks. Most pioneering studies either learn from only a single teacher in their distillation learning methods, neglecting the potential that a student can learn from multiple teachers simultaneously, or simply treat each teacher to be equally important, unable to reveal the different importance of teachers for specific examples. To bridge this gap, we propose a novel adaptive multi-teacher multi-level knowledge distillation learning framework (AMTML-KD), which consists two novel insights: (i) associating each teacher with a latent representation to adaptively learn instance-level teacher importance weights which are leveraged for acquiring integrated soft-targets (high-level knowledge) and (ii) enabling the intermediate-level hints (intermediate-level knowledge) to be gathered from multiple teachers by the proposed multi-group hint strategy. As such, a student model can learn multi-level knowledge from multiple teachers through AMTML-KD. Extensive results on publicly available datasets demonstrate the proposed learning framework ensures student to achieve improved performance than strong competitors.
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.
Integrity signals
Registry
Cites
Statistics
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.
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
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
A fast learning algorithm for deep belief nets
Hinton, G. E., Osindero, S., & Teh, Y.-W. (2006) · Neural Computation · doi:10.1162/neco.2006.18.7.1527 · Cited by 16,336
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204
Ensemble methods in machine learning
Dietterich, T. G. (2000) · Lecture Notes in Computer Science · doi:10.1007/3-540-45014-9_1 · Cited by 8,408
FitNets: Hints for Thin Deep Nets
Romero, A., Ballas, N., Kahou, S. E., Chassang, A., Gatta, C., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.6550 · Cited by 2,859
Buciluǎ, C., Caruana, R., & Niculescu-Mizil, A. (2006) · Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1150402.1150464 · Cited by 2,068
Unifying distillation and privileged information
Lopez-Paz, D., Bottou, L., Schölkopf, B., & Vapnik, V. (2015) · arXiv · doi:10.48550/arxiv.1511.03643 · Cited by 270 · cited 1×
“Note a similar learning strategy, privilege learning [25], focuses on using privilege information only existing in training stage to help optimizing models and is unified with distillation learning [13].”
Background
Research works printed but not linked: 23
- Ba, 2014, Do deep nets really need to be deep?, NIPS, 2654
- Y. Bachrach, Y. Finkelstein, R. Gilad-Bachrach, L. Katzir, N. Koenigstein, N. Nice, U. Paquet, Speeding up the xbox recommender system using a euclidean transformation for inner-product spaces, in: RecSys, ACM, 2014, pp. 257–264 · DOI · Paper page
- Fukuda, 2017, Efficient knowledge distillation from an ensemble of teachers, Interspeech, 3697
- Z. Hu, X. Ma, Z. Liu, E. Hovy, E. Xing, Harnessing deep neural networks with logic rules, 2016. arXiv preprint arXiv:1603.06318. · DOI · Paper page
- Y. Koren, Factorization meets the neighborhood: a multifaceted collaborative filtering model, in: SIGKDD, ACM, 2008, pp. 426–434 · DOI · Paper page
- Liu, 2019, Structured knowledge distillation for semantic segmentation, CVPR, 2604
- Mun, 2018, Learning to specialize with knowledge distillation for visual question answering, NIPS, 8092
- Nakashole, 2017, Knowledge distillation for bilingual dictionary induction, EMNLP, 2497
- Nam, 2017, Dual attention networks for multimodal reasoning and matching, CVPR, 299
- Park, 2019, Relational knowledge distillation, CVPR, 3967
- S. Ruder, P. Ghaffari, J.G. Breslin, Knowledge adaptation: Teaching to adapt, in: ICLR, 2017.
- Russakovsky, 2015, Imagenet large scale visual recognition challenge, IJCV, 115, 211, 10.1007/s11263-015-0816-y · DOI · Paper page
- B.B. Sau, V.N. Balasubramanian, V.N., Deep model compression: Distilling knowledge from noisy teachers, 2016. arXiv preprint arXiv:1610.09650. · arXiv
- K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, 2014. arXiv preprint arXiv:1409.1556. · arXiv
- X. Song, F. Feng, X. Han, X. Yang, W. Liu, L. Nie, Neural compatibility modeling with attentive knowledge distillation, in: SIGIR, ACM, 2018, pp. 5–14. · DOI
- X. Tan, Y. Ren, D. He, T. Qin, Z. Zhao, T. Liu, Multilingual neural machine translation with knowledge distillation, in: ICLR, 2019. · DOI
- Vapnik, 2015, Learning using privileged information: similarity control and knowledge transfer, JMLR, 16, 2
- Vongkulbhisal, 2019, Unifying heterogeneous classifiers with distillation, CVPR, 3175
- Yim, 2017, A gift from knowledge distillation: fast optimization, network minimization and transfer learning, CVPR, 4133
- You, 2017, Learning from multiple teacher networks, SIGKDD, 1285
- You, 2018, Learning with single-teacher multi-student, AAAI, 4390
- S. Zagoruyko, N. Komodakis, Paying more attention to attention: improving the performance of convolutional neural networks via attention transfer. CoRR abs/1612.03928, 2016. URL:http://arxiv.org/abs/1612.03928, arXiv:1612.03928. · arXiv
- Zhang, 2018, Deep mutual learning, CVPR, 4320
Cited by
Knowledge Distillation: A Survey
Gou, J., Yu, B., Maybank, S. J., & Tao, D. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01453-z · Cited by 3,931
Bringing ai to edge: from deep learning’s perspective
Liu, D., Kong, H., Luo, X., Liu, W., & Subramaniam, R. (2022) · Neurocomputing · doi:10.1016/j.neucom.2021.04.141 · Cited by 171
Metadata
- Authors
- Liu, Y., Zhang, W., & Wang, J.
- Venue
- Neurocomputing
- Publisher
- Elsevier BV
- Published
- 2020-11
- Volume
- 415
- Pages
- 106-113
- ISSN
- 0925-2312 (ISSN-L)
- OpenAlex
- W3045672834
- Cited by
- 221
- References
- 8
- FWCI
- 7.98
- Type
- journal article
- Language
- en
- Open access
- green
- Licence
- www.elsevier.com/tdm/userlicense/1.0/ (tdm)
- www.elsevier.com/legal/tdmrep-license (tdm)
- Funders
- National Natural Science Foundation of China
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: Adaptive multi-teacher multi-level knowledge distillation. https://app.cochat.ai/papers/10.1016/j.neucom.2020.07.048
- The paper
- Liu, Y., Zhang, W., & Wang, J. (2020). Adaptive multi-teacher multi-level knowledge distillation. Neurocomputing. https://doi.org/10.1016/j.neucom.2020.07.048
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.