Pith. sign in

REVIEW 2 major objections 5 minor 1 cited by

Energy-Aware Deep Learning on Resource-Constrained Hardware

T0 review · 2 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This survey argues that energy consumption is a distinct optimization axis for deep learning on constrained hardware, one that MAC and FLOP counts do not capture, and that the field still lacks accurate hardware-agnostic energy estimation.

desk verdict Useful survey of energy-aware DL on constrained devices, but the motivating SqueezeNet claim and the NEq equation need fixing before this is reliable. read the letter →

arxiv 2505.12523 v1 pith:MBTB6DEK submitted 2025-05-18 cs.LG cs.AR

Josh Millar , Hamed Haddadi , Anil Madhavapeddy This is my paper · ORCID
classification cs.LGcs.AR
keywords energy-awaredeeplearningTinyMLon-deviceinferenceenergyestimationneuralarchitecturesearchmulti-exitnetworksharvestingfederated
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This survey argues that energy use is a first-class constraint for deep learning on IoT and mobile hardware, separate from memory and compute budgets. Its central case is that conventional efficiency proxies—multiply-accumulate (MAC) and FLOP counts—do not track energy consumption, because data movement, not arithmetic, dominates the energy bill and because memory hierarchies differ across devices. The paper organizes the field into energy-aware architecture design, adaptive inference (right-sizing, early exits, offloading), on-device training, and energy-harvesting or federated deployment, and it identifies accurate hardware-agnostic energy estimation as the unsolved problem on which the whole area depends.

What carries the argument

The load-bearing mechanism is the data-movement cost model: fetching data from memory far from the compute unit can cost $10$ to $100+$ times an arithmetic operation, and feature-map movement rather than computation is what dominates DNN energy. This model is what turns the proxy-failure observation into a design principle: since energy depends on where data lives and how it flows, any energy-aware method must estimate or measure per-layer, per-platform movement costs rather than counting operations.

What would settle it

Measure the end-to-end energy of SqueezeNet and AlexNet on the same microcontroller while controlling for batch and input resolution: the survey's motivating claim predicts SqueezeNet can consume more energy despite having far fewer multiply-accumulate operations, so a dataset of platforms where energy tracks MAC count monotonically would falsify the proxy-failure premise.

Watch

Extended reading notes

Core claim

The authors claim that energy-aware deep learning is a distinct optimization axis that today's MAC- or FLOP-based proxies systematically miss, and they support this with evidence that a network with far fewer operations can consume more energy on a given platform, and that only about 10% of a typical CNN's energy goes to computation while the rest goes to moving feature maps. They therefore classify existing work by how it makes energy itself the objective: pruning and quantization guided by layer-wise energy estimates, neural architecture search that predicts energy from measurements or regressions, inference policies that trade accuracy for energy at runtime, and training or fine-tuning methods that limit which parameters are updated. The survey concludes that the field's binding constraint is the lack of a universal, execution-free way to estimate a DNN's energy on arbitrary hardware, and proposes architecture representations such as abstract syntax trees as a route toward hardware-agnostic estimation.

Load-bearing premise

The survey's taxonomy and recommendations assume that its summaries of the cited works are accurate, including quantitative details such as SqueezeNet's '50x fewer MACs' comparison and NEq's equilibrium condition; the paper itself contains at least one misstatement (SqueezeNet's actual cited result is 50x fewer parameters, not MACs) and a misprinted NEq inequality.

Editorial extensions

If this is right

  • Energy-aware pruning and quantization must be evaluated on measured energy, not on MAC or parameter counts, or they may silently increase consumption.
  • Adaptive inference—early exits, input-dependent quantization, and offloading—becomes the main lever for staying inside a fluctuating energy budget on battery-free devices.
  • On-device fine-tuning for data shifts will need parameter-selection or rehearsal methods, since full backpropagation is too energy-expensive on microcontrollers.
  • Federated learning over heterogeneous, intermittently powered devices needs energy-aware participation policies, or the global model becomes biased toward well-powered devices.
  • Progress on any of these fronts is gated by the same missing capability: a cheap, accurate, execution-free energy estimator that works across hardware.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If cross-platform energy estimation is ever solved, compiler-style cost models for DNNs could make energy-awareness an automated part of the build, much as latency is today.
  • The data-movement emphasis predicts that the best compression recipe for one memory hierarchy will not transfer to another, so per-device calibration may be unavoidable even with a hardware-agnostic core model.
  • Energy-aware neural architecture search with uncertainty-aware energy prediction could avoid overfitting to the few benchmarked devices and make search practical for microcontroller-scale deployment.
  • Intermittent energy-harvesting machine learning, if made reliable, would let remote sensors run vision or audio classifiers for years without batteries, changing where on-device AI is economically sensible.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. This survey reviews energy-aware deep learning techniques for resource-constrained IoT and mobile hardware. It organizes the field into energy-aware DNN design (pruning, quantization, neural architecture search), energy-adaptive inference (right-sizing, multi-exit networks, offloading), on-device training, and applications such as energy-harvesting systems and federated learning. The paper argues that energy consumption is not captured by MAC/FLOP proxies and that accurate, hardware-agnostic energy estimation remains an open problem, concluding with a set of future research directions and tables summarizing existing methods, estimation approaches, and embedded ML frameworks.

Significance. If the survey's summaries are accurate, it provides a valuable synthesis of a fragmented literature and a useful starting point for researchers entering this area. The organizational scheme, the coverage of intermittent computing and on-device training, and the tables of NAS methods, energy-estimation approaches, and MCU frameworks are concrete contributions. However, the survey's value as a reference depends on the fidelity of its descriptions of primary sources; the misattributed SqueezeNet example in §2 and the uninterpretable NEq equation in §4 undermine confidence in the cited summaries and must be corrected. The paper also demonstrates areas that are rarely surveyed together, such as energy-harvesting and federated learning, which gives it a distinct niche among existing reviews.

major comments (2)
  1. [§2, first paragraph] The motivating example for why MAC proxies fail to predict energy is misattributed: the paper states that 'SqueezeNet [51] contains 50x fewer MACs than AlexNet [68], yet exhibits greater energy consumption on various platforms [103, 173].' Reference [51] (Iandola et al., SqueezeNet) reports 50x fewer parameters and a model size below 1MB, not 50x fewer MACs. The cited references do not clearly establish the comparative energy-consumption claim on 'various platforms.' Because this example is the central evidence for the paper's thesis that MAC counts are not an energy proxy, the authors should replace it with a documented MAC/energy comparison or rephrase the claim to match the sources.
  2. [§4, Eq. (2)] The equilibrium condition for NEq is uninterpretable as written: the inequality '|v_t_i| < epsilon, epsilon <= 0' is impossible for the non-negative magnitude |v_t_i|. The following sentence mentions beta ('β is an arbitrary threshold'), but β is not defined in Eq. (2) or the surrounding text, and the text also refers to 'epsilon/beta' as if the two are interchangeable. Readers cannot determine NEq's actual selection rule from this description. The authors should reproduce NEq's exact condition with all symbols defined, or remove the equation and explain the method in words.
minor comments (5)
  1. [§2, footnote] The orphaned footnote '0https://www.st.com/resource/en/datasheet/stm32l4r5zi.pdf' at the bottom of page 2 should be removed or converted into a proper citation in the text.
  2. [Reference [150]] The YOLOv5 reference contains the unfinished placeholder 'Accessed: insert date here.' and needs to be completed before publication.
  3. [Reference list] Several distinct reference numbers point to the same work (e.g., [125] and [126]; [173] and [174]; [107] and [108]; [143] and [144]; [55] and [56]; [175] and [176]; [49] and [50]; [39] and [67]). Deduplicate these entries and renumber consistently so that readers can trace claims to unique sources.
  4. [§3.3, Eq. (1)] The weight α in the class-dependent threshold equation is not defined in the text; please state its meaning, typical range, and whether it is a hyperparameter set by the user.
  5. [§4, text after Eq. (2)] The sentence introducing β appears to be a leftover from a different version of the manuscript, since Eq. (2) uses only ε. Define β and explain how it relates to ε, or remove the mention.

Circularity Check

0 steps flagged · score 1.0 of 10

Survey is a literature synthesis with no derivation chain; the single self-citation is not load-bearing, so circularity is minimal.

full rationale

This paper is a survey, not a derivation: it organizes and summarizes existing energy-aware DL methods, and it makes no formal prediction that could reduce to its inputs. The central claims are taxonomic and descriptive, and the only self-citation, reference [188] (co-authored by Haddadi), appears once in Section 5.1 as an example of underwater bioacoustics monitoring and does not support any load-bearing argument, taxonomy, or future-work recommendation. No parameter is fitted, no quantity is defined in terms of another claimed result, and no uniqueness or existence theorem is imported from the authors' prior work. The identified concerns in the manuscript, such as the SqueezeNet '50x fewer MACs' statement and the uninterpretable NEq equilibrium condition in Section 4, are factual-fidelity and correctness risks in summarizing cited work, not instances of circular reasoning: a summary can be wrong without the survey's conclusions being definitionally equivalent to its sources. Because the paper presents no derivation chain and its only self-citation is peripheral, the honest finding is no significant circularity, with a score of 1 reflecting the negligible presence of a non-load-bearing self-citation rather than any circular derivation.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

This survey presents no new derivation or data; its substance consists of organizing external literature. The main unverified premises are the accuracy of the cited summaries and the reliability of the hardware energy measurements reported by the original papers. Two parameters appear in quoted formulas without proper definition, which we list as free parameters because the survey itself presents them.

free parameters (2)
  • alpha (energy weight) in Eq. (1) = not specified
    The class-dependent threshold formula includes α as a weighting between false-positive rate and energy, but the survey never defines α or suggests a value.
  • epsilon/beta (equilibrium threshold) in Eq. (2) = not specified
    The NEq equilibrium condition is printed as |v| < ε with ε ≤ 0, which is impossible; the text then uses β. Neither constant is defined.
assumptions (3)
  • domain assumption The cited works are accurately represented in the survey's summaries.
    The survey does not reproduce measurements or verify the original papers; all conclusions depend on these summaries being correct.
  • domain assumption Reported hardware energy measurements from cited papers are reliable.
    The survey's comparisons of methods across platforms rely on the accuracy of measurements that are not independently checked.
  • ad hoc to paper The definition of 'energy-aware DL' (in §1) marks a coherent and useful category.
    The authors define the scope by fiat, excluding works that optimize only latency or memory; the value of the survey depends on this categorization being useful.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Energy-Aware Deep Learning on Resource-Constrained Hardware." pith.science (2026). https://pith.science/paper/MBTB6DEK

@misc{pith2026250512523,
  author       = {Pith},
  title        = {Pith review of: Energy-Aware Deep Learning on Resource-Constrained Hardware},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MBTB6DEK}},
  note         = {Machine review of arXiv:2505.12523}
}
read the original abstract

The use of deep learning (DL) on Internet of Things (IoT) and mobile devices offers numerous advantages over cloud-based processing. However, such devices face substantial energy constraints to prolong battery-life, or may even operate intermittently via energy-harvesting. Consequently, \textit{energy-aware} approaches for optimizing DL inference and training on such resource-constrained devices have garnered recent interest. We present an overview of such approaches, outlining their methodologies, implications for energy consumption and system-level efficiency, and their limitations in terms of supported network types, hardware platforms, and application scenarios. We hope our review offers a clear synthesis of the evolving energy-aware DL landscape and serves as a foundation for future research in energy-constrained computing.

Figures

Figures reproduced from arXiv: 2505.12523 by the authors.

Figure 1
Figure 1. Conventional DNN architecture alongside an EE variant. The dashed lines indicate conditional execution paths. Another design decision is the granularity and positioning of exit layers. Coarse-grained exits reduce overheads, but can miss exit opportunities. Conversely, fine-grained exits increase overheads, leading to heightened energy requirements and increased parameters. The extent of these overheads depends on th… view at source ↗
Figure 2
Figure 2. Multi-exit last-layer distillation training, with a backbone network trained in the cloud and its exits trained on-device. prediction entropy [143], and score margin (SM), the margin between the 1 𝑠𝑡 and 2 𝑛𝑑 inference output scores [108, 110]. However, imposing a global threshold on network outputs corresponds to implicitly assuming all classes are equally difficult to process. In scenarios for which this assumptio… view at source ↗
Figure 3
Figure 3. DNN partitioning between device and server. 4 On-Device Training The above sections generally focus on optimizing on-device inference with an offline-trained network. However once deployed, data- or context-shifts can result in degraded network performance in the target environment [135]. This issue is especially relevant for in-the-wild sensing applications, where environmental conditions can be highly variable and… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. FOGNITE: Federated Learning-Enhanced Fog-Cloud Architecture

    cs.DC 2025-07 reject novelty 4.0 of 10

    FOGNITE combines federated learning, RL, and digital twins for smart grid load balancing, but its reported gains are inconsistent between abstract and body.

Reference graph

Works this paper leans on

184 extracted references · 24 canonical work pages · cited by 1 Pith paper

  1. [108]

    Priyadarshini Panda, Abhronil Sengupta, and Kaushik Roy. 2015. Conditional Deep Learning for Energy-Efficient and Enhanced Pattern Recognition. CoRR abs/1509.08971 (2015). arXiv:1509.08971 http://arxiv.org/abs/1509.08971

    arXiv 2015

  2. [174]

    Tien-Ju Yang, Yu-Hsin Chen, and Vivienne Sze. 2016. Designing Energy-Efficient Convolutional Neural Networks using Energy-Aware Pruning. CoRR abs/1611.05128 (2016). arXiv:1611.05128 http://arxiv.org/abs/1611.05128

    arXiv 2016

  3. [50]

    Weinberger

    Gao Huang, Danlu Chen, Tianhong Li, Felix Wu, Laurens van der Maaten, and Kilian Q. Weinberger. 2017. Multi-Scale Dense Convolutional Networks for Efficient Prediction. CoRR abs/1703.09844 (2017). arXiv:1703.09844 http://arxiv.org/abs/1703.09844

    arXiv 2017

  4. [51]

    Iandola, Matthew W

    Forrest N. Iandola, Matthew W. Moskewicz, Khalid Ashraf, Song Han, William J. Dally, and Kurt Keutzer. 2016. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size.CoRR abs/1602.07360 (2016). arXiv:1602.07360 http://arxiv.org/abs/1602.07360

    arXiv 2016

  5. [68]

    Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. 2012. ImageNet Classification with Deep Convolutional Neural Networks. In Advances in Neural Information Processing Systems , F. Pereira, C.J. Burges, L. Bottou, and K.Q. Weinberger (Eds.), Vol. 25. Curran Associates, Inc. https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c84...

    2012

  6. [1]

    Youssef Abadade, Anas Temouden, Hatim Bamoumen, Nabil Benamar, Yousra Chtouki, and Abdelhakim Senhaji Hafid. 2023. A Comprehensive Survey on TinyML. IEEE Access 11 (2023), 96892–96922. doi:10.1109/ACCESS.2023.3294111

    arXiv 2023

  7. [2]

    Anders Andreassen, Yasaman Bahri, Behnam Neyshabur, and Rebecca Roelofs. 2021. The Evolution of Out-of-Distribution Robustness Throughout Fine-Tuning. CoRR abs/2106.15831 (2021). arXiv:2106.15831 https://arxiv.org/abs/2106.15831

    arXiv 2021

  8. [3]

    Apache. 2024. MicroTVM. https://tvm.apache.org/docs/topic/microtvm/index.html

    2024

Show all 184 references
  1. [4]

    ARM. 2024. CMSIS-NN. https://github.com/ARM-software/CMSIS-NN

    2024

  2. [5]

    Pedram Bakhtiarifard, Christian Igel, and Raghavendra Selvan. 2024. EC-NAS: Energy Consumption Aware Tabular Benchmarks for Neural Architecture Search. arXiv:2210.06015 [cs.LG]

    2024 arXiv

  3. [6]

    Yogesh Balaji, Swami Sankaranarayanan, and Rama Chellappa. 2018. MetaReg: Towards Domain Generalization using Meta- Regularization. In Advances in Neural Information Processing Systems , S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa- Bianchi, and R. Garnett (Eds.),...

    2018

  4. [7]

    Konstantin Berestizshevsky and Guy Even. 2018. Sacrificing Accuracy for Reduced Computation: Cascaded Inference Based on Softmax Confidence. CoRR abs/1805.10982 (2018). arXiv:1805.10982 http://arxiv.org/abs/1805.10982

    2018 arXiv

  5. [8]

    Andrea Bragagnolo, Enzo Tartaglione, and Marco Grangetto. 2022. To update or not to update? Neurons at equilibrium in deep models. arXiv:2207.09455 [cs.LG]

    2022 arXiv

  6. [9]

    Han Cai, Ji Lin, Yujun Lin, Zhijian Liu, Haotian Tang, Hanrui Wang, Ligeng Zhu, and Song Han. 2022. Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications. ACM Transactions on Design Automation of Electronic Systems 27, 3 (March 2022), 1–50. doi:10.1145/3486618

    2022 doi

  7. [10]

    Han Cai, Ligeng Zhu, and Song Han. 2018. ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware. CoRR abs/1812.00332 (2018). arXiv:1812.00332 http://arxiv.org/abs/1812.00332

    2018 arXiv

  8. [11]

    Tianqi Chen, Bing Xu, Chiyuan Zhang, and Carlos Guestrin. 2016. Training Deep Nets with Sublinear Memory Cost. arXiv:1604.06174 [cs.LG]

    2016 arXiv

  9. [12]

    Yanxi Chen, Xuchen Pan, Yaliang Li, Bolin Ding, and Jingren Zhou. 2024. EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism. arXiv:2312.04916 [cs.LG]

    2024 arXiv

  10. [13]

    Yang Chen, Xiaoyan Sun, and Yaochu Jin. 2020. Communication-Efficient Federated Deep Learning With Layerwise Asynchronous Model Update and Temporally Weighted Aggregation. IEEE Transactions on Neural Networks and Learning Systems 31, 10 (Oct. 2020), 4229–4238. doi:10.1109/tnnl...

    2020

  11. [14]

    Yu-Hsin Chen, Joel Emer, and Vivienne Sze. 2016. Eyeriss: A Spatial Architecture for Energy-Efficient Dataflow for Convolutional Neural Networks. In 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA) . 367–379. doi:10.1109/ISCA.2016.40

    2016 doi

  12. [15]

    Alexei Colin and Brandon Lucia. 2016. Chain: tasks and channels for reliable intermittent programs. In Proceedings of the 2016 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications (Amsterdam, Netherlands) (OOPSLA 2016). Asso...

    2016

  13. [16]

    Francesco Daghero, Alessio Burrello, Daniele Jahier Pagliari, Luca Benini, Enrico Macii, and Massimo Poncino. 2020. Energy-Efficient Adaptive Machine Learning on IoT End-Nodes With Class-Dependent Confidence. In 2020 27th IEEE International Conference on Electronics, Circuits ...

    2020

  14. [17]

    Francesco Daghero, Alessio Burrello, Chen Xie, Marco Castellano, Luca Gandolfi, Andrea Calimera, Enrico Macii, Massimo Poncino, and Daniele Jahier Pagliari. 2022. Human Activity Recognition on Microcontrollers with Quantized and Adaptive Deep Neural Networks. ACM Transactions ...

    2022 doi

  15. [18]

    Xiaoliang Dai, Peizhao Zhang, Bichen Wu, Hongxu Yin, Fei Sun, Yanghan Wang, Marat Dukhan, Yunqing Hu, Yiming Wu, Yangqing Jia, Peter Vajda, Matt Uyttendaele, and Niraj K. Jha. 2019. ChamNet: Towards Efficient Network Design Through Platform-Aware Model Adaptation. In Proceedin...

    2019

  16. [19]

    Carmen Delgado and Jeroen Famaey. 2022. Optimal Energy-Aware Task Scheduling for Batteryless IoT Devices. IEEE Transactions on Emerging Topics in Computing 10, 3 (2022), 1374–1387. doi:10.1109/TETC.2021.3086144

    2022

  17. [20]

    Swarnava Dey, Arijit Mukherjee, Arpan Pal, and Balamuralidhar P. 2019. Embedded Deep Inference in Practice: Case for Model Partitioning. In Proceedings of the 1st Workshop on Machine Learning on Edge in Sensor Systems (New York, NY, USA) (SenSys-ML 2019). Association for Compu...

    2019

  18. [21]

    Dong Dong, Hongxu Jiang, Xuekai Wei, Yanfei Song, Xu Zhuang, and Jason Wang. 2023. ETNAS: An energy consumption task-driven neural architecture search. Sustainable Computing: Informatics and Systems 40 (2023), 100926. doi:10.1016/j.suscom.2023.100926

    2023

  19. [22]

    Xuanyi Dong and Yi Yang. 2020. NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search. arXiv:2001.00326 [cs.CV]

    2020 arXiv

  20. [23]

    Ying-Jun Du, Jun Xu, Huan Xiong, Qiang Qiu, Xiantong Zhen, Cees G. M. Snoek, and Ling Shao. 2020. Learning to Learn with Variational Information Bottleneck for Domain Generalization. CoRR abs/2007.07645 (2020). arXiv:2007.07645 https://arxiv.org/abs/2007.07645 Energy-Aware Dee...

    2020 arXiv

  21. [24]

    I (Still) Can’t Believe It’s Not Better!

    Cian Eastwood, Ian Mason, and Christopher K. I. Williams. 2022. Unit-level surprise in neural networks. InProceedings on "I (Still) Can’t Believe It’s Not Better!" at NeurIPS 2021 Workshops (Proceedings of Machine Learning Research, Vol. 163) , Melanie F. Pradier, Aaron Schein...

    2022

  22. [25]

    Amir Erfan Eshratifar, Mohammad Saeed Abrishami, and Massoud Pedram. 2018. JointDNN: An Efficient Training and Inference Engine for Intelligent Mobile Cloud Computing Services. CoRR abs/1801.08618 (2018). arXiv:1801.08618 http://arxiv.org/abs/1801.08618

    2018 arXiv

  23. [26]

    Biyi Fang, Xiao Zeng, Faen Zhang, Hui Xu, and Mi Zhang. 2020. FlexDNN: Input-Adaptive On-Device Deep Learning for Efficient Mobile Vision. In 2020 IEEE/ACM Symposium on Edge Computing (SEC) . 84–95. doi:10.1109/SEC50012.2020.00014

    2020

  24. [27]

    Terry Fawden, Lorena Qendro, and Cecilia Mascolo. 2023. Uncertainty-Informed On-Device Personalisation Using Early Exit Networks on Sensor Signals. 2023 31st European Signal Processing Conference (EUSIPCO) (2023), 1305–1309. https://api.semanticscholar.org/ CorpusID:261116963

    2023

  25. [28]

    Collins, Yukun Zhu, Li Zhang, Jonathan Huang, Dmitry P

    Michael Figurnov, Maxwell D. Collins, Yukun Zhu, Li Zhang, Jonathan Huang, Dmitry P. Vetrov, and Ruslan Salakhutdinov. 2016. Spatially Adaptive Computation Time for Residual Networks. CoRR abs/1612.02297 (2016). arXiv:1612.02297 http://arxiv.org/abs/1612. 02297

    2016 arXiv

  26. [29]

    Francesco Fraternali, Bharathan Balaji, Yuvraj Agarwal, and Rajesh K. Gupta. 2020. ACES: Automatic Configuration of Energy Harvesting Sensors with Reinforcement Learning. ACM Transactions on Sensor Networks 16, 4 (July 2020), 1–31. doi:10.1145/3404191

    2020 doi

  27. [30]

    Mullins, and Cheng-Zhong Xu

    Xitong Gao, Yiren Zhao, Lukasz Dudziak, Robert D. Mullins, and Cheng-Zhong Xu. 2018. Dynamic Channel Pruning: Feature Boosting and Suppression. CoRR abs/1810.05331 (2018). arXiv:1810.05331 http://arxiv.org/abs/1810.05331

    2018 arXiv

  28. [31]

    In Gim and JeongGil Ko. 2022. Memory-efficient DNN training on mobile devices. In Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services (Portland, Oregon) (MobiSys ’22). Association for Computing Machinery, New York, NY, USA, 464–...

    2022

  29. [32]

    Marco Giordano, Philipp Mayer, and Michele Magno. 2020. A Battery-Free Long-Range Wireless Smart Camera for Face Detection. In Proceedings of the 8th International Workshop on Energy Harvesting and Energy-Neutral Sensing Systems (Virtual Event, Japan) (ENSsys ’20). Association...

    2020

  30. [33]

    Graham Gobieski, Nathan Beckmann, and Brandon Lucia. 2018. Intelligence Beyond the Edge: Inference on Intermittent Embedded Systems. CoRR abs/1810.07751 (2018). arXiv:1810.07751 http://arxiv.org/abs/1810.07751

    2018 arXiv

  31. [34]

    Google. 2024. TFLite-Micro. https://github.com/tensorflow/tflite-micro

    2024

  32. [35]

    Basak Guler and Aylin Yener. 2021. Sustainable Federated Learning. CoRR abs/2102.11274 (2021). arXiv:2102.11274 https://arxiv.org/ abs/2102.11274

    2021 arXiv

  33. [36]

    Yunhui Guo, Honghui Shi, Abhishek Kumar, Kristen Grauman, Tajana Rosing, and Rogério Schmidt Feris. 2018. SpotTune: Transfer Learning through Adaptive Fine-tuning. CoRR abs/1811.08737 (2018). arXiv:1811.08737 http://arxiv.org/abs/1811.08737

    2018 arXiv

  34. [37]

    Pengchao Han, Shiqiang Wang, and Kin K. Leung. 2020. Adaptive Gradient Sparsification for Efficient Federated Learning: An Online Learning Approach. CoRR abs/2001.04756 (2020). arXiv:2001.04756 https://arxiv.org/abs/2001.04756

    2020 arXiv

  35. [39]

    Song Han, Huizi Mao, and William J. Dally. 2016. Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding. arXiv:1510.00149 [cs.CV]

    2016 arXiv

  36. [40]

    Song Han, Jeff Pool, John Tran, and William J. Dally. 2015. Learning both Weights and Connections for Efficient Neural Networks. CoRR abs/1506.02626 (2015). arXiv:1506.02626 http://arxiv.org/abs/1506.02626

    2015 arXiv

  37. [41]

    Mohammad Hasan. 2022. State of IoT-Spring 2022. IOT Analytics, See https://iot-analytics. com/product/state-of-iot-spring-2022 website (2022)

    2022

  38. [43]

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2015. Deep Residual Learning for Image Recognition. arXiv:1512.03385 [cs.CV]

    2015 arXiv

  39. [44]

    Josiah Hester, Kevin Storer, and Jacob Sorber. 2017. Timely Execution on Intermittently Powered Batteryless Sensors. 1–13. doi:10. 1145/3131672.3131673

    2017

  40. [45]

    Matthew Hicks. 2017. Clank: Architectural support for intermittent computation. In 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA) . 228–240. doi:10.1145/3079856.3080238

    2017

  41. [46]

    Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam

    Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. 2017. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv:1704.04861 [cs.CV]

    2017 arXiv

  42. [47]

    Hanpeng Hu, Junwei Su, Juntao Zhao, Yanghua Peng, Yibo Zhu, Haibin Lin, and Chuan Wu. 2024. CDMPP: A Device-Model Agnostic Framework for Latency Prediction of Tensor Programs. In Proceedings of the Nineteenth European Conference on Computer Systems (EuroSys ’24). ACM. doi:10.1...

    2024

  43. [48]

    Edward Suh

    Weizhe Hua, Christopher De Sa, Zhiru Zhang, and G. Edward Suh. 2018. Channel Gating Neural Networks. CoRR abs/1805.12549 (2018). arXiv:1805.12549 http://arxiv.org/abs/1805.12549 18 • Millar et al

    2018 arXiv

  44. [52]

    Bashima Islam and Shahriar Nirjon. 2020. Zygarde: Time-Sensitive On-Device Deep Inference and Adaptation on Intermittently-Powered Systems. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 4, 3, Article 82 (sep 2020), 29 pages. doi:10.1145/3411808

    2020 doi

  45. [53]

    Yesmina Jaafra, Jean Luc Laurent, Aline Deruyver, and Mohamed Saber Naceur. 2019. Reinforcement learning for neural architecture search: A review. Image and Vision Computing 89 (2019), 57–66. doi:10.1016/j.imavis.2019.06.005

    2019 doi

  46. [54]

    Daniele Jahier Pagliari, Francesco Daghero, and Massimo Poncino. 2020. Sequence-To-Sequence Neural Networks Inference on Embedded Processors Using Dynamic Beam Search. Electronics 9, 2 (2020). doi:10.3390/electronics9020337

    2020 doi

  47. [56]

    Daniele Jahier Pagliari, Francesco Panini, Enrico Macii, and Massimo Poncino. 2019. Dynamic Beam Width Tuning for Energy-Efficient Recurrent Neural Networks. In Proceedings of the 2019 on Great Lakes Symposium on VLSI (Tysons Corner, VA, USA)(GLSVLSI ’19). Association for Comp...

    2019

  48. [57]

    Seunghyeok Jeon, Yonghun Choi, Yeonwoo Cho, and Hojung Cha. 2023. HarvNet: Resource-Optimized Operation of Multi-Exit Deep Neural Networks on Energy Harvesting Devices. In Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services (Hel...

    2023

  49. [58]

    Guan, and Maya Gupta

    Heinrich Jiang, Been Kim, Melody Y. Guan, and Maya Gupta. 2018. To Trust Or Not To Trust A Classifier. arXiv:1805.11783 [stat.ML]

    2018 arXiv

  50. [59]

    Petar Jokic, Stephane Emery, and Luca Benini. 2021. Battery-Less Face Recognition at the Extreme Edge. In 2021 19th IEEE International New Circuits and Systems Conference (NEWCAS) . 1–4. doi:10.1109/NEWCAS50681.2021.9462787

    2021

  51. [60]

    Beomseok Kang, Anni Lu, Yun Long, Daehyun Kim, Shimeng Yu, and Saibal Mukhopadhyay. 2021. Genetic Algorithm-Based Energy- Aware CNN Quantization for Processing-In-Memory Architecture. IEEE Journal on Emerging and Selected Topics in Circuits and Systems 11, 4 (2021), 649–662. d...

    2021

  52. [61]

    Yiping Kang, Johann Hauswald, Cao Gao, Austin Rovinski, Trevor Mudge, Jason Mars, and Lingjia Tang. 2017. Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge. SIGPLAN Not. 52, 4 (apr 2017), 615–629. doi:10.1145/3093336.3037698

    2017

  53. [62]

    Mohsen Karimi, Hyunjong Choi, Yidi Wang, Yecheng Xiang, and Hyoseung Kim. 2021. Real-Time Task Scheduling on Intermittently Powered Batteryless Devices. IEEE Internet of Things Journal 8, 17 (2021), 13328–13342. doi:10.1109/JIOT.2021.3065947

    2021

  54. [63]

    Yigitcan Kaya and Tudor Dumitras. 2018. How to Stop Off-the-Shelf Deep Neural Networks from Overthinking. CoRR abs/1810.07052 (2018). arXiv:1810.07052 http://arxiv.org/abs/1810.07052

    2018 arXiv

  55. [64]

    Aria Khoshsirat, Giovanni Perin, and Michele Rossi. 2023. Divide and Save: Splitting Workload Among Containers in an Edge Device to Save Energy and Time. arXiv:2302.06478 [cs.DC]

    2023

  56. [65]

    Kulkarni, and Tony Tae-Hyoung Kim

    Donghyuk Kim, Chengshuo Yu, Shanshan Xie, Yuzong Chen, Joo-Young Kim, Bongjin Kim, Jaydeep P. Kulkarni, and Tony Tae-Hyoung Kim. 2022. An Overview of Processing-in-Memory Circuits for Artificial Intelligence and Machine Learning. IEEE Journal on Emerging and Selected Topics in...

    2022

  57. [66]

    Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A

    James Kirkpatrick, Razvan Pascanu, Neil C. Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A. Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, Demis Hassabis, Claudia Clopath, Dharshan Kumaran, and Raia Hadsell. 2016. Overcoming catastrophic for...

    2016 arXiv

  58. [67]

    Raghuraman Krishnamoorthi. 2018. Quantizing deep convolutional networks for efficient inference: A whitepaper. arXiv:1806.08342 [cs.LG]

    2018 arXiv

  59. [69]

    Ananya Kumar, Aditi Raghunathan, Robbie Jones, Tengyu Ma, and Percy Liang. 2022. Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution. arXiv:2202.10054 [cs.LG]

    2022 arXiv

  60. [70]

    Kwon, Jagmohan Chauhan, Hong Jia, Stylianos I

    Young D. Kwon, Jagmohan Chauhan, Hong Jia, Stylianos I. Venieris, and Cecilia Mascolo. 2023. LifeLearner: Hardware-Aware Meta Continual Learning System for Embedded Computing Platforms. arXiv:2311.11420 [cs.LG]

    2023 arXiv

  61. [71]

    Kwon, Rui Li, Stylianos I

    Young D. Kwon, Rui Li, Stylianos I. Venieris, Jagmohan Chauhan, Nicholas D. Lane, and Cecilia Mascolo. 2023. TinyTrain: Deep Neural Network Training at the Extreme Edge. arXiv:2307.09988 [cs.LG]

    2023 arXiv

  62. [72]

    Stefanos Laskaridis, Alexandros Kouris, and Nicholas D. Lane. 2021. Adaptive Inference through Early-Exit Networks: Design, Challenges and Directions. In Proceedings of the 5th International Workshop on Embedded and Mobile Deep Learning (Virtual, WI, USA) Energy-Aware Deep Lea...

    2021

  63. [73]

    Venieris, Hyeji Kim, and Nicholas D

    Stefanos Laskaridis, Stylianos I. Venieris, Hyeji Kim, and Nicholas D. Lane. 2020. HAPI: Hardware-Aware Progressive Inference. CoRR abs/2008.03997 (2020). arXiv:2008.03997 https://arxiv.org/abs/2008.03997

    2020 arXiv

  64. [74]

    Jaejun Lee, Raphael Tang, and Jimmy Lin. 2019. What Would Elsa Do? Freezing Layers During Transformer Fine-Tuning. CoRR abs/1911.03090 (2019). arXiv:1911.03090 http://arxiv.org/abs/1911.03090

    2019 arXiv

  65. [75]

    Seulki Lee, Bashima Islam, Yubo Luo, and Shahriar Nirjon. 2020. Intermittent Learning: On-Device Machine Learning on Intermittently Powered System. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 3, 4, Article 141 (sep 2020), 30 pages. doi:10.1145/3369837

    2020 doi

  66. [76]

    Chen, Fahim Tajwar, Ananya Kumar, Huaxiu Yao, Percy Liang, and Chelsea Finn

    Yoonho Lee, Annie S. Chen, Fahim Tajwar, Ananya Kumar, Huaxiu Yao, Percy Liang, and Chelsea Finn. 2023. Surgical Fine-Tuning Improves Adaptation to Distribution Shifts. arXiv:2210.11466 [cs.LG]

    2023 arXiv

  67. [77]

    Venieris, and Nicholas D

    Ilias Leontiadis, Stefanos Laskaridis, Stylianos I. Venieris, and Nicholas D. Lane. 2021. It’s always personal: Using Early Exits for Efficient On-Device CNN Personalisation. CoRR abs/2102.01393 (2021). arXiv:2102.01393 https://arxiv.org/abs/2102.01393

    2021 arXiv

  68. [78]

    Chaojian Li, Zhongzhi Yu, Yonggan Fu, Yongan Zhang, Yang Zhao, Haoran You, Qixuan Yu, Yue Wang, and Yingyan Lin. 2021. HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark. CoRR abs/2103.10584 (2021). arXiv:2103.10584 https: //arxiv.org/abs/2103.10584

    2021 arXiv

  69. [79]

    Hospedales

    Da Li, Yongxin Yang, Yi-Zhe Song, and Timothy M. Hospedales. 2017. Learning to Generalize: Meta-Learning for Domain Generalization. arXiv:1710.03463 [cs.LG]

    2017 arXiv

  70. [80]

    Xiang Li, Kaixuan Huang, Wenhao Yang, Shusen Wang, and Zhihua Zhang. 2020. On the Convergence of FedAvg on Non-IID Data. arXiv:1907.02189 [stat.ML]

    2020 arXiv

  71. [81]

    Xiangjie Li, Chenfei Lou, Zhengping Zhu, Yuchi Chen, Yingtao Shen, Yehan Ma, and An Zou. 2022. Predictive Exit: Prediction of Fine-Grained Early Exits for Computation- and Energy-Efficient Inference. arXiv:2206.04685 [cs.LG]

    2022 arXiv

  72. [82]

    Edgar Liberis, Lukasz Dudziak, and Nicholas D. Lane. 2020. 𝜇NAS: Constrained Neural Architecture Search for Microcontrollers. CoRR abs/2010.14246 (2020). arXiv:2010.14246 https://arxiv.org/abs/2010.14246

    2020 arXiv

  73. [83]

    Ji Lin, Wei-Ming Chen, Yujun Lin, John Cohn, Chuang Gan, and Song Han. 2020. MCUNet: Tiny Deep Learning on IoT Devices. CoRR abs/2007.10319 (2020). arXiv:2007.10319 https://arxiv.org/abs/2007.10319

    2020 arXiv

  74. [84]

    Ji Lin, Yongming Rao, Jiwen Lu, and Jie Zhou. 2017. Runtime Neural Pruning. In Advances in Neural Information Processing Systems , I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.), Vol. 30. Curran Associates, Inc. https://proce...

    2017

  75. [85]

    Ji Lin, Ligeng Zhu, Wei-Ming Chen, Wei-Chen Wang, Chuang Gan, and Song Han. 2024. On-Device Training Under 256KB Memory. arXiv:2206.15472 [cs.CV]

    2024 arXiv

  76. [86]

    Qiang Liu, Lemeng Wu, and Dilin Wang. 2019. Splitting Steepest Descent for Growing Neural Architectures. CoRR abs/1910.02366 (2019). arXiv:1910.02366 http://arxiv.org/abs/1910.02366

    2019 arXiv

  77. [87]

    Weijie Liu, Xiaoxi Zhang, Jingpu Duan, Carlee Joe-Wong, Zhi Zhou, and Xu Chen. 2023. AdaCoOpt: Leverage the Interplay of Batch Size and Aggregation Frequency for Federated Learning. In 2023 IEEE/ACM 31st International Symposium on Quality of Service (IWQoS) . 1–10. doi:10.1109...

    2023

  78. [88]

    Weijie Liu, Xiaoxi Zhang, Jingpu Duan, Carlee Joe-Wong, Zhi Zhou, and Xu Chen. 2023. DYNAMITE: Dynamic Interplay of Mini-Batch Size and Aggregation Frequency for Federated Learning with Static and Streaming Dataset. arXiv:2310.14906 [cs.LG]

    2023 arXiv

  79. [89]

    Yuhan Liu, Saurabh Agarwal, and Shivaram Venkataraman. 2021. AutoFreeze: Automatically Freezing Model Blocks to Accelerate Fine-tuning. CoRR abs/2102.01386 (2021). arXiv:2102.01386 https://arxiv.org/abs/2102.01386

    2021 arXiv

  80. [90]

    Yuqiao Liu, Yanan Sun, Bing Xue, Mengjie Zhang, and Gary G. Yen. 2020. A Survey on Evolutionary Neural Architecture Search. CoRR abs/2008.10937 (2020). arXiv:2008.10937 https://arxiv.org/abs/2008.10937

    2020 arXiv

  81. [91]

    Bing Luo, Xiang Li, Shiqiang Wang, Jianwei Huang, and Leandros Tassiulas. 2020. Cost-Effective Federated Learning Design. CoRR abs/2012.08336 (2020). arXiv:2012.08336 https://arxiv.org/abs/2012.08336

    2020 arXiv

  82. [92]

    Kaisheng Ma, Xueqing Li, Jinyang Li, Yongpan Liu, Yuan Xie, Jack Sampson, Mahmut Taylan Kandemir, and Vijaykrishnan Narayanan

  83. [93]

    Kaisheng Ma, Yang Zheng, Shuangchen Li, Karthik Swaminathan, Xueqing Li, Yongpan Liu, Jack Sampson, Yuan Xie, and Vijaykrishnan Narayanan. 2015. Architecture exploration for ambient energy harvesting nonvolatile processors. In 2015 IEEE 21st International Symposium on High Per...

    2015

  84. [94]

    Ningning Ma, Xiangyu Zhang, Hai-Tao Zheng, and Jian Sun. 2018. ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design. arXiv:1807.11164 [cs.CV]

    2018 arXiv

  85. [95]

    Divyat Mahajan, Shruti Tople, and Amit Sharma. 2020. Domain Generalization using Causal Matching. CoRR abs/2006.07500 (2020). arXiv:2006.07500 https://arxiv.org/abs/2006.07500

    2020 arXiv

  86. [96]

    Andrea Maioli and Luca Mottola. 2021. ALFRED: Virtual Memory for Intermittent Computing. InProceedings of the 19th ACM Conference on Embedded Networked Sensor Systems (SenSys ’21) . ACM. doi:10.1145/3485730.3485949 20 • Millar et al

    2021

  87. [97]

    Milone, and Enzo Ferrante

    Lucas Mansilla, Rodrigo Echeveste, Diego H. Milone, and Enzo Ferrante. 2021. Domain Generalization via Gradient Surgery. CoRR abs/2108.01621 (2021). arXiv:2108.01621 https://arxiv.org/abs/2108.01621

    2021 arXiv

  88. [98]

    Alberto Marchisio, Andrea Massa, Vojtech Mrazek, Beatrice Bussolino, Maurizio Martina, and Muhammad Shafique. 2020. NASCaps: A Framework for Neural Architecture Search to Optimize the Accuracy and Hardware Efficiency of Convolutional Capsule Networks. CoRR abs/2008.08476 (2020...

    2020 arXiv

  89. [99]

    Brendan McMahan, Eider Moore, Daniel Ramage, and Blaise Agüera y Arcas

    H. Brendan McMahan, Eider Moore, Daniel Ramage, and Blaise Agüera y Arcas. 2016. Federated Learning of Deep Networks using Model Averaging. CoRR abs/1602.05629 (2016). arXiv:1602.05629 http://arxiv.org/abs/1602.05629

    2016 arXiv

  90. [100]

    Lingchen Meng, Hengduo Li, Bor-Chun Chen, Shiyi Lan, Zuxuan Wu, Yu-Gang Jiang, and Ser-Nam Lim. 2021. AdaViT: Adaptive Vision Transformers for Efficient Image Recognition. CoRR abs/2111.15668 (2021). arXiv:2111.15668 https://arxiv.org/abs/2111.15668

    2021 arXiv

  91. [101]

    Mateusz Michalkiewicz, Masoud Faraki, Xiang Yu, Manmohan Chandraker, and Mahsa Baktashmotlagh. 2023. Domain Generalization Guided by Gradient Signal to Noise Ratio of Parameters. arXiv:2310.07361 [cs.CV]

    2023 arXiv

  92. [102]

    Microsoft. 2024. EdgeML. https://microsoft.github.io/EdgeML/

    2024

  93. [103]

    The cost of everything and the value of nothing

    David Moloney. 2016. Embedded deep neural networks: “The cost of everything and the value of nothing”. In 2016 IEEE Hot Chips 28 Symposium (HCS). 1–20. doi:10.1109/HOTCHIPS.2016.7936219

    2016

  94. [104]

    Alessandro Montanari, Manuja Sharma, Dainius Jenkus, Mohammed Alloulah, Lorena Qendro, and Fahim Kawsar. 2020. ePerceptive: energy reactive embedded intelligence for batteryless sensors. In Proceedings of the 18th Conference on Embedded Networked Sensor Systems (Virtual Event,...

    2020 doi

  95. [105]

    Augustus Odena, Dieterich Lawson, and Christopher Olah. 2017. Changing Model Behavior at Test-Time Using Reinforcement Learning. arXiv:1702.07780 [stat.ML]

    2017 arXiv

  96. [106]

    Daniele Jahier Pagliari, Enrico Macii, and Massimo Poncino. 2018. Dynamic Bit-width Reconfiguration for Energy-Efficient Deep Learning Hardware. In Proceedings of the International Symposium on Low Power Electronics and Design (Seattle, WA, USA) (ISLPED ’18). Association for C...

    2018

  97. [109]

    Yanwei Pang, Haoran Wang, Yunlong Yu, and Zhong Ji. 2019. A decadal survey of zero-shot image classification. SCIENTIA SINICA Informationis (2019). https://api.semanticscholar.org/CorpusID:208104980

    2019

  98. [110]

    Eunhyeok Park, Dongyoung Kim, Soobeom Kim, Yong-Deok Kim, Gunhee Kim, Sungroh Yoon, and Sungjoo Yoo. 2015. Big/little deep neural network for ultra low power inference. In 2015 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS). 124–132. d...

    2015

  99. [111]

    Christos Profentzas, Magnus Almgren, and Olaf Landsiedel. 2023. MiniLearn: On-Device Learning for Low-Power IoT Devices. In Proceedings of the 2022 International Conference on Embedded Wireless Systems and Networks (, Linz, Austria,) (EWSN ’22). Association for Computing Machi...

    2023

  100. [112]

    Aël Quélennec, Enzo Tartaglione, Pavlo Mozharovskyi, and Van-Tam Nguyen. 2023. Towards On-device Learning on the Edge: Ways to Select Neurons to Update under a Budget Constraint. arXiv:2312.05282 [cs.LG]

    2023 arXiv

  101. [113]

    Jathushan Rajasegaran, Vinoj Jayasundara, Sandaru Jayasekara, Hirunima Jayasekara, Suranga Seneviratne, and Ranga Rodrigo. 2019. DeepCaps: Going Deeper with Capsule Networks. arXiv:1904.09546 [cs.CV]

    2019 arXiv

  102. [114]

    Ramasesh, Ethan Dyer, and Maithra Raghu

    Vinay V. Ramasesh, Ethan Dyer, and Maithra Raghu. 2020. Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics. CoRR abs/2007.07400 (2020). arXiv:2007.07400 https://arxiv.org/abs/2007.07400

    2020 arXiv

  103. [115]

    Mohammad Rastegari, Vicente Ordonez, Joseph Redmon, and Ali Farhadi. 2016. XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks. arXiv:1603.05279 [cs.CV]

    2016 arXiv

  104. [116]

    Pengzhen Ren, Yun Xiao, Xiaojun Chang, Po-Yao Huang, Zhihui Li, Xiaojiang Chen, and Xin Wang. 2020. A Comprehensive Survey of Neural Architecture Search: Challenges and Solutions. CoRR abs/2006.02903 (2020). arXiv:2006.02903 https://arxiv.org/abs/2006.02903

    2020 arXiv

  105. [117]

    Crefeda Faviola Rodrigues, Graham Riley, and Mikel Lujan. 2018. Fine-Grained Energy and Performance Profiling framework for Deep Convolutional Neural Networks. arXiv:1803.11151 [cs.PF]

    2018 arXiv

  106. [118]

    Bita Darvish Rouhani, Azalia Mirhoseini, and Farinaz Koushanfar. 2016. DeLight: Adding Energy Dimension To Deep Neural Networks. In Proceedings of the 2016 International Symposium on Low Power Electronics and Design (San Francisco Airport, CA, USA) (ISLPED ’16). Association fo...

    2016

  107. [119]

    Amelie Royer and Christoph H. Lampert. 2020. A Flexible Selection Scheme for Minimum-Effort Transfer Learning.CoRR abs/2008.11995 (2020). arXiv:2008.11995 https://arxiv.org/abs/2008.11995

    2020 arXiv

  108. [120]

    Sara Sabour, Nicholas Frosst, and Geoffrey E. Hinton. 2017. Dynamic Routing Between Capsules. CoRR abs/1710.09829 (2017). arXiv:1710.09829 http://arxiv.org/abs/1710.09829 Energy-Aware Deep Learning on Resource-Constrained Hardware • 21

    2017 arXiv

  109. [121]

    Adnan Sabovic, Michiel Aernouts, Dragan Subotic, Jaron Fontaine, Eli De Poorter, and Jeroen Famaey. 2023. Towards energy-aware tinyML on battery-less IoT devices. Internet of Things 22 (2023), 100736. doi:10.1016/j.iot.2023.100736

    2023

  110. [122]

    Adnan Sabovic, Ashish Kumar Sultania, Carmen Delgado, Lander De Roeck, and Jeroen Famaey. 2022. An Energy-Aware Task Scheduler for Energy-Harvesting Batteryless IoT Devices. IEEE Internet of Things Journal 9, 22 (2022), 23097–23114. doi:10.1109/JIOT.2022.3185321

    2022

  111. [123]

    Hadi Salman, Andrew Ilyas, Logan Engstrom, Ashish Kapoor, and Aleksander Madry. 2020. Do Adversarially Robust ImageNet Models Transfer Better? CoRR abs/2007.08489 (2020). arXiv:2007.08489 https://arxiv.org/abs/2007.08489

    2020 arXiv

  112. [124]

    Eric Samikwa, Antonio Di Maio, and Torsten Braun. 2022. Adaptive Early Exit of Computation for Energy-Efficient and Low-Latency Machine Learning over IoT Networks. In 2022 IEEE 19th Annual Consumer Communications & Networking Conference (CCNC) . 200–206. doi:10.1109/CCNC49033....

    2022

  113. [126]

    Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. 2019. MobileNetV2: Inverted Residuals and Linear Bottlenecks. arXiv:1801.04381 [cs.CV]

    2019 arXiv

  114. [127]

    Felix Sattler, Simon Wiedemann, Klaus-Robert Müller, and Wojciech Samek. 2019. Robust and Communication-Efficient Federated Learning from Non-IID Data. CoRR abs/1903.02891 (2019). arXiv:1903.02891 http://arxiv.org/abs/1903.02891

    2019 arXiv

  115. [128]

    Mark, and Ravi Teja Mullapudi

    Noam Shazeer, Kayvon Fatahalian, William R. Mark, and Ravi Teja Mullapudi. 2018. HydraNets: Specialized Dynamic Architectures for Efficient Inference. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . 8080–8089. doi:10.1109/CVPR.2018.00843

    2018

  116. [129]

    Le, Geoffrey E

    Noam Shazeer, Azalia Mirhoseini, Krzysztof Maziarz, Andy Davis, Quoc V. Le, Geoffrey E. Hinton, and Jeff Dean. 2017. Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer. CoRR abs/1701.06538 (2017). arXiv:1701.06538 http://arxiv.org/ abs/1701.06538

    2017 arXiv

  117. [130]

    Paras Sheth and Huan Liu. 2023. Causal Domain Generalization. Springer International Publishing, Cham, 161–185. doi:10.1007/978-3- 031-35051-1_8

    2023 doi

  118. [131]

    Selçuk Candan, Adrienne Raglin, and Huan Liu

    Paras Sheth, Raha Moraffah, K. Selçuk Candan, Adrienne Raglin, and Huan Liu. 2022. Domain Generalization – A Causal Perspective. arXiv:2209.15177 [cs.LG]

    2022 arXiv

  119. [132]

    Dian Shi, Liang Li, Rui Chen, Pavana Prakash, Miao Pan, and Yuguang Fang. 2022. Toward Energy-Efficient Federated Learning Over 5G+ Mobile Devices. IEEE Wireless Communications 29, 5 (2022), 44–51. doi:10.1109/MWC.003.2100028

    2022 doi

  120. [133]

    Dian Shi, Liang Li, Maoqiang Wu, Minglei Shu, Rong Yu, Miao Pan, and Zhu Han. 2022. To Talk or to Work: Dynamic Batch Sizes Assisted Time Efficient Federated Learning Over Future Mobile Edge Devices. IEEE Transactions on Wireless Communications 21, 12 (2022), 11038–11050. doi:...

    2022

  121. [134]

    Yuge Shi, Jeffrey Seely, Philip H. S. Torr, N. Siddharth, Awni Y. Hannun, Nicolas Usunier, and Gabriel Synnaeve. 2021. Gradient Matching for Domain Generalization. CoRR abs/2104.09937 (2021). arXiv:2104.09937 https://arxiv.org/abs/2104.09937

    2021 arXiv

  122. [135]

    Spadaro, R

    G. Spadaro, R. Renzulli, A. Bragagnolo, J. H. Giraldo, A. Fiandrotti, M. Grangetto, and E. Tartaglione. 2023. Shannon Strikes Again! Entropy-based Pruning in Deep Neural Networks for Transfer Learning under Extreme Memory and Computation Budgets. In 2023 IEEE/CVF International...

    2023

  123. [136]

    STMicroelectronics. 2024. STM32CubeMX. https://www.st.com/en/development-tools/stm32cubemx.html

    2024

  124. [137]

    Ashish Kumar Sultania and Jeroen Famaey. 2022. Batteryless Bluetooth Low Energy Prototype With Energy-Aware Bidirectional Communication Powered by Ambient Light. IEEE Sensors Journal 22, 7 (2022), 6685–6697. doi:10.1109/JSEN.2022.3153097

    2022

  125. [138]

    Yuxuan Sun, Sheng Zhou, and Deniz Gündüz. 2019. Energy-Aware Analog Aggregation for Federated Learning with Redundant Data. arXiv:1911.00188 [cs.IT]

    2019 arXiv

  126. [139]

    Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich

    Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott E. Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. 2014. Going Deeper with Convolutions. CoRR abs/1409.4842 (2014). arXiv:1409.4842 http://arxiv.org/abs/1409.4842

    2014 arXiv

  127. [140]

    Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, and Quoc V. Le. 2018. MnasNet: Platform-Aware Neural Architecture Search for Mobile. CoRR abs/1807.11626 (2018). arXiv:1807.11626 http://arxiv.org/abs/1807.11626

    2018 arXiv

  128. [141]

    Mingxing Tan and Quoc V. Le. 2019. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. CoRR abs/1905.11946 (2019). arXiv:1905.11946 http://arxiv.org/abs/1905.11946

    2019 arXiv

  129. [142]

    Iris Bahar, and Sherief Reda

    Hokchhay Tann, Soheil Hashemi, R. Iris Bahar, and Sherief Reda. 2016. Runtime Configurable Deep Neural Networks for Energy- Accuracy Trade-off. CoRR abs/1607.05418 (2016). arXiv:1607.05418 http://arxiv.org/abs/1607.05418

    2016 arXiv

  130. [144]

    Surat Teerapittayanon, Bradley McDanel, and H. T. Kung. 2017. BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks. CoRR abs/1709.01686 (2017). arXiv:1709.01686 http://arxiv.org/abs/1709.01686

    2017 arXiv

  131. [145]

    Nazli Tekin, Ahmet Aris, Abbas Acar, Selcuk Uluagac, and Vehbi Cagri Gungor. 2024. A review of on-device machine learning for IoT: An energy perspective. Ad Hoc Networks 153 (2024), 103348. doi:10.1016/j.adhoc.2023.103348 22 • Millar et al

    2024

  132. [146]

    Tran, Wei Bao, Albert Zomaya, Minh N

    Nguyen H. Tran, Wei Bao, Albert Zomaya, Minh N. H. Nguyen, and Choong Seon Hong. 2019. Federated Learning over Wireless Networks: Optimization Model Design and Analysis. InIEEE INFOCOM 2019 - IEEE Conference on Computer Communications. 1387–1395. doi:10.1109/INFOCOM.2019.8737464

    2019

  133. [147]

    Hoang Truong, Shuo Zhang, Ufuk Muncuk, Phuc Nguyen, Nam Bui, Anh Nguyen, Qin Lv, Kaushik Chowdhury, Thang Dinh, and Tam Vu. 2018. CapBand: Battery-free Successive Capacitance Sensing Wristband for Hand Gesture Recognition. In Proceedings of the 16th ACM Conference on Embedded ...

    2018

  134. [148]

    John, Arjun Suresh, Rowan Taubitz, Sean Zhan, Scott Wasson, David Kanter, and Vijay Janapa Reddi

    Arya Tschand, Arun Tejusve Raghunath Rajan, Sachin Idgunji, Anirban Ghosh, Jeremy Holleman, Csaba Kiraly, Pawan Ambalkar, Ritika Borkar, Ramesh Chukka, Trevor Cockrell, Oliver Curtis, Grigori Fursin, Miro Hodak, Hiwot Kassa, Anton Lokhmotov, Dejan Miskovic, Yuechao Pan, Manu P...

    2025 arXiv

  135. [149]

    X. Tu, A. Mallik, D. Chen, K. Han, O. Altintas, H. Wang, and J. Xie. 2023. Unveiling Energy Efficiency in Deep Learning: Measurement, Prediction, and Scoring Across Edge Devices. In 2023 IEEE/ACM Symposium on Edge Computing (SEC) . IEEE Computer Society, Los Alamitos, CA, USA,...

    2023

  136. [150]

    Ultralytics. 2021. YOLOv5: A state-of-the-art real-time object detection system. https://docs.ultralytics.com. Accessed: insert date here

    2021

  137. [151]

    uTensor. 2024. uTensor. https://github.com/uTensor/uTensor

    2024

  138. [152]

    Shubham Vaishnav, Maria Efthymiou, and Sindri Magnússon. 2023. Energy-Efficient and Adaptive Gradient Sparsification for Federated Learning. In ICC 2023 - IEEE International Conference on Communications . 1256–1261. doi:10.1109/ICC45041.2023.10278999

    2023

  139. [153]

    Joel Van Der Woude and Matthew Hicks. 2016. Intermittent computation without hardware support or programmer intervention. In Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation (Savannah, GA, USA) (OSDI’16). USENIX Association, USA, 17–32

    2016

  140. [154]

    Philipp van Kempen, Rafael Stahl, Daniel Mueller-Gritschneder, and Ulf Schlichtmann. 2023. MLonMCU: TinyML Benchmarking with Fast Retargeting. In Proceedings of the 2023 Workshop on Compilers, Deployment, and Tooling for Edge AI (CODAI ’23) . ACM. doi:10.1145/3615338.3618128

    2023

  141. [155]

    Dilin Wang, Meng Li, Lemeng Wu, Vikas Chandra, and Qiang Liu. 2019. Energy-Aware Neural Architecture Optimization with Fast Splitting Steepest Descent. CoRR abs/1910.03103 (2019). arXiv:1910.03103 http://arxiv.org/abs/1910.03103

    2019 arXiv

  142. [156]

    Qipeng Wang, Mengwei Xu, Chao Jin, Xinran Dong, Jinliang Yuan, Xin Jin, Gang Huang, Yunxin Liu, and Xuanzhe Liu. 2022. Melon: breaking the memory wall for resource-efficient on-device machine learning. In Proceedings of the 20th Annual International Conference on Mobile System...

    2022

  143. [157]

    Leung, Christian Makaya, Ting He, and Kevin Chan

    Shiqiang Wang, Tiffany Tuor, Theodoros Salonidis, Kin K. Leung, Christian Makaya, Ting He, and Kevin Chan. 2019. Adaptive Federated Learning in Resource Constrained Edge Computing Systems. IEEE Journal on Selected Areas in Communications 37, 6 (2019), 1205–1221. doi:10.1109/JS...

    2019

  144. [158]

    Zheng, Han Yu, and Chunyan Miao

    Wei Wang, Vincent W. Zheng, Han Yu, and Chunyan Miao. 2019. A Survey of Zero-Shot Learning: Settings, Methods, and Applications. ACM Trans. Intell. Syst. Technol. 10, 2, Article 13 (jan 2019), 37 pages. doi:10.1145/3293318

    2019 doi

  145. [159]

    Gonzalez

    Xin Wang, Fisher Yu, Zi-Yi Dou, and Joseph E. Gonzalez. 2017. SkipNet: Learning Dynamic Routing in Convolutional Networks. CoRR abs/1711.09485 (2017). arXiv:1711.09485 http://arxiv.org/abs/1711.09485

    2017 arXiv

  146. [160]

    Yulin Wang, Rui Huang, Shiji Song, Zeyi Huang, and Gao Huang. 2021. Not All Images are Worth 16x16 Words: Dynamic Vision Transformers with Adaptive Sequence Length. CoRR abs/2105.15075 (2021). arXiv:2105.15075 https://arxiv.org/abs/2105.15075

    2021 arXiv

  147. [161]

    Christopher J. C. H. Watkins and Peter Dayan. 1992. Q-learning.Machine Learning 8, 3 (01 May 1992), 279–292. doi:10.1007/BF00992698

    1992 doi

  148. [162]

    Kang Wei, Jun Li, Chuan Ma, Ming Ding, Feng Shu, Haitao Zhao, Wen Chen, and Hongbo Zhu. 2024. Gradient sparsification for efficient wireless federated learning with differential privacy. Science China Information Sciences 67, 4 (March 2024). doi:10.1007/s11432-023- 3918-9

    2024 doi

  149. [163]

    Taylan Cemgil

    Olivia Wiles, Sven Gowal, Florian Stimberg, Sylvestre-Alvise Rebuffi, Ira Ktena, Krishnamurthy Dvijotham, and A. Taylan Cemgil

  150. [164]

    R. J. Williams. 1992. Simple statistical gradient-following algorithms for connectionist reinforcement learning. Machine Learning 8 (1992), 229–256

    1992

  151. [165]

    Podolak, Jacek Tabor, Marek Smieja, and Tomasz Trzcinski

    Maciej Wolczyk, Bartosz Wójcik, Klaudia Balazy, Igor T. Podolak, Jacek Tabor, Marek Smieja, and Tomasz Trzcinski. 2021. Zero Time Waste: Recycling Predictions in Early Exit Neural Networks. CoRR abs/2106.05409 (2021). arXiv:2106.05409 https://arxiv.org/abs/2106. 05409

    2021 arXiv

  152. [166]

    Bichen Wu, Xiaoliang Dai, Peizhao Zhang, Yanghan Wang, Fei Sun, Yiming Wu, Yuandong Tian, Peter Vajda, Yangqing Jia, and Kurt Keutzer. 2019. FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search. arXiv:1812.03443 [cs.CV]

    2019 arXiv

  153. [167]

    Davis, Kristen Grauman, and Rogerio Feris

    Zuxuan Wu, Tushar Nagarajan, Abhishek Kumar, Steven Rennie, Larry S. Davis, Kristen Grauman, and Rogerio Feris. 2019. BlockDrop: Dynamic Inference Paths in Residual Networks. arXiv:1711.08393 [cs.CV] Energy-Aware Deep Learning on Resource-Constrained Hardware • 23

    2019 arXiv

  154. [168]

    Sang Michael Xie, Ananya Kumar, Robbie Jones, Fereshte Khani, Tengyu Ma, and Percy Liang. 2020. In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness. CoRR abs/2012.04550 (2020). arXiv:2012.04550 https: //arxiv.org/abs/2012.04550

    2020 arXiv

  155. [169]

    Ji Xin, Raphael Tang, Jaejun Lee, Yaoliang Yu, and Jimmy Lin. 2020. DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference. CoRR abs/2004.12993 (2020). arXiv:2004.12993 https://arxiv.org/abs/2004.12993

    2020 arXiv

  156. [170]

    Ji Xin, Raphael Tang, Yaoliang Yu, and Jimmy J. Lin. 2021. BERxiT: Early Exiting for BERT with Better Fine-Tuning and Extension to Regression. In Conference of the European Chapter of the Association for Computational Linguistics . https://api.semanticscholar.org/ CorpusID:233189542

    2021

  157. [171]

    Mengwei Xu, Feng Qian, Mengze Zhu, Feifan Huang, Saumay Pushp, and Xuanzhe Liu. 2020. DeepWear: Adaptive Local Offloading for On-Wearable Deep Learning. IEEE Transactions on Mobile Computing 19, 2 (2020), 314–330. doi:10.1109/TMC.2019.2893250

    2020

  158. [172]

    Fan Yang, Ashok Samraj Thangarajan, Gowri Sankar Ramachandran, Wouter Joosen, and Danny Hughes. 2021. AsTAR: Sustainable Energy Harvesting for the Internet of Things through Adaptive Task Scheduling. ACM Trans. Sen. Netw. 18, 1, Article 4 (oct 2021), 34 pages. doi:10.1145/3467894

    2021 doi

  159. [176]

    Tien-Ju Yang, Yu-Hsin Chen, Joel Emer, and Vivienne Sze. 2017. A method to estimate the energy consumption of deep neural networks. In 2017 51st Asilomar Conference on Signals, Systems, and Computers . 1916–1920. doi:10.1109/ACSSC.2017.8335698

    2017

  160. [177]

    Zhaohui Yang, Mingzhe Chen, Walid Saad, Choong Seon Hong, and Mohammad Shikh-Bahaei. 2019. Energy Efficient Federated Learning Over Wireless Communication Networks. CoRR abs/1911.02417 (2019). arXiv:1911.02417 http://arxiv.org/abs/1911.02417

    2019 arXiv

  161. [178]

    Shuochao Yao, Jinyang Li, Dongxin Liu, Tianshi Wang, Shengzhong Liu, Huajie Shao, and Tarek Abdelzaher. 2020. Deep compressive offloading: speeding up neural network inference by trading edge computation for network latency. InProceedings of the 18th Conference on Embedded Net...

    2020

  162. [179]

    Seul-Ki Yeom, Kyung-Hwan Shim, and Jee-Hyun Hwang. 2021. Toward Compact Deep Neural Networks via Energy-Aware Pruning. CoRR abs/2103.10858 (2021). arXiv:2103.10858 https://arxiv.org/abs/2103.10858

    2021 arXiv

  163. [180]

    Amirreza Yousefzadeh, Jan Stuijt, Martijn Hijdra, Hsiao-Hsuan Liu, Anteneh Gebregiorgis, Abhairaj Singh, Said Hamdioui, and Francky Catthoor. 2022. Energy-efficient In-Memory Address Calculation. ACM Trans. Archit. Code Optim. 19, 4, Article 52 (sep 2022), 16 pages. doi:10.114...

    2022 doi

  164. [181]

    Jiahui Yu, Linjie Yang, Ning Xu, Jianchao Yang, and Thomas S. Huang. 2018. Slimmable Neural Networks. CoRR abs/1812.08928 (2018). arXiv:1812.08928 http://arxiv.org/abs/1812.08928

    2018 arXiv

  165. [182]

    Rong Yu and Peichun Li. 2021. Toward Resource-Efficient Federated Learning in Mobile Edge Computing. IEEE Network 35, 1 (2021), 148–155. doi:10.1109/MNET.011.2000295

    2021 doi

  166. [183]

    Jinliang Yuan, Shangguang Wang, Hongyu Li, Daliang Xu, Yuanchun Li, Mengwei Xu, and Xuanzhe Liu. 2024. Towards Energy-efficient Federated Learning via INT8-based Training on Mobile DSPs. In Proceedings of the ACM Web Conference 2024 (Singapore, Singapore) (WWW ’24). Associatio...

    2024

  167. [184]

    Arber Zela, Julien Siems, and Frank Hutter. 2020. NAS-Bench-1Shot1: Benchmarking and Dissecting One-shot Neural Architecture Search. CoRR abs/2001.10422 (2020). arXiv:2001.10422 https://arxiv.org/abs/2001.10422

    2020 arXiv

  168. [185]

    Liekang Zeng, En Li, Zhi Zhou, and Xu Chen. 2019. Boomerang: On-Demand Cooperative Deep Neural Network Inference for Edge Intelligence on the Industrial Internet of Things. IEEE Network 33, 5 (2019), 96–103. doi:10.1109/MNET.001.1800506

    2019 doi

  169. [186]

    Leung, and Kaibin Huang

    Qunsong Zeng, Yuqing Du, Kin K. Leung, and Kaibin Huang. 2019. Energy-Efficient Radio Resource Allocation for Federated Edge Learning. arXiv:1907.06040 [cs.IT]

    2019 arXiv

  170. [187]

    Yu Zhang, Tao Gu, and Xi Zhang. 2022. MDLdroidLite: A Release-and-Inhibit Control Approach to Resource-Efficient Deep Neural Networks on Mobile Devices. IEEE Transactions on Mobile Computing 21, 10 (2022), 3670–3686. doi:10.1109/TMC.2021.3062575

    2022

  171. [188]

    Yuchen Zhao, Sayed Saad Afzal, Waleed Akbar, Osvy Rodriguez, Fan Mo, David Boyle, Fadel Adib, and Hamed Haddadi. 2022. Towards battery-free machine learning and inference in underwater environments. In Proceedings of the 23rd Annual International Workshop on Mobile Computing S...

    2022

  172. [189]

    Yiyang Zhao, Linnan Wang, Yuandong Tian, Rodrigo Fonseca, and Tian Guo. 2020. Few-shot Neural Architecture Search. CoRR abs/2006.06863 (2020). arXiv:2006.06863 https://arxiv.org/abs/2006.06863

    2020 arXiv

  173. [190]

    Zhuoran Zhao, Kamyar Mirzazad Barijough, and Andreas Gerstlauer. 2018. DeepThings: Distributed Adaptive Deep Learning Inference on Resource-Constrained IoT Edge Clusters. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 37, 11 (2018), 2348–2359. do...

    2018

  174. [191]

    Guangxu Zhu, Yong Wang, and Kaibin Huang. 2020. Broadband Analog Aggregation for Low-Latency Federated Edge Learning. IEEE Transactions on Wireless Communications 19, 1 (2020), 491–506. doi:10.1109/TWC.2019.2946245

    2020

  175. [2017]

    In2017 50th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO)

    Incidental Computing on IoT Nonvolatile Processors. In2017 50th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO). 204–218

  176. [2021]

    CoRR abs/2110.11328 (2021)

    A Fine-Grained Analysis on Distribution Shift. CoRR abs/2110.11328 (2021). arXiv:2110.11328 https://arxiv.org/abs/2110.11328

    2021 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.