mlsanityData to 14 Oct 2026

LifeLearner: Hardware-Aware Meta Continual Learning System for Embedded Computing Platforms

arXiv 2311.11420 ↗First posted 19 Nov 2023cs.LG cs.AI cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

Continual Learning (CL) allows applications such as user personalization and household robots to learn on the fly and adapt to context. This is an important feature when context, actions, and users change. However, enabling CL on resource-constrained embedded systems is challenging due to the limited labeled data, memory, and computing capacity. In this paper, we propose LifeLearner, a hardware-aware meta continual learning system that drastically optimizes system resources (lower memory, latency, energy consumption) while ensuring high accuracy. Specifically, we (1) exploit meta-learning and rehearsal strategies to explicitly cope with data scarcity issues and ensure high accuracy, (2) effectively combine lossless and lossy compression to significantly reduce the resource requirements of CL and rehearsal samples, and (3) developed hardware-aware system on embedded and IoT platforms considering the hardware characteristics. As a result, LifeLearner achieves near-optimal CL performance, falling short by only 2.8% on accuracy compared to an Oracle baseline. With respect to the state-of-the-art (SOTA) Meta CL method, LifeLearner drastically reduces the memory footprint (by 178.7x), end-to-end latency by 80.8-94.2%, and energy consumption by 80.9-94.2%. In addition, we successfully deployed LifeLearner on two edge devices and a microcontroller unit, thereby enabling efficient CL on resource-constrained platforms where it would be impractical to run SOTA methods and the far-reaching deployment of adaptable CL in a ubiquitous manner. Code is available at https://github.com/theyoungkwon/LifeLearner.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 61 of its 106 references that are in the map. First posted on arXiv 19 Nov 2023.

Builds on

61 of 106 references.

References · 61

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  5. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  6. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  7. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  8. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  9. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  10. 2017Billion-scale similarity search with GPUscited by 2,497
  11. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,333
  12. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
38 more
  1. 2016Matching Networks for One Shot Learningcited by 2,255
  2. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  3. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  4. 2016Progressive Neural Networkscited by 1,361
  5. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  6. 2018ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Designcited by 1,136
  7. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  8. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 849
  9. 2016Training Deep Nets with Sublinear Memory Costcited by 848
  10. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 752
  11. 2019Large Scale Incremental Learningcited by 732
  12. 2017Lifelong Learning with Dynamically Expandable Networkscited by 694
  13. 2018End-to-End Incremental Learningcited by 597
  14. 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 521
  15. 2018Progress & Compress: A scalable framework for continual learningcited by 520
  16. 2017Efficient Processing of Deep Neural Networks: A Tutorial and Surveycited by 434
  17. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  18. 2019Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examplescited by 315
  19. 2019Compacting, Picking and Growing for Unforgetting Continual Learningcited by 207
  20. 2019REMIND Your Neural Network to Prevent Catastrophic Forgettingcited by 197
  21. 2019Meta-Learning Representations for Continual Learningcited by 189
  22. 2020TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systemscited by 165
  23. 2022On-Device Training Under 256KB Memorycited by 160
  24. 2020Training with Quantization Noise for Extreme Model Compressioncited by 145
  25. 2022Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Differencecited by 126
  26. 2017Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environmentscited by 108
  27. 2019Latent Replay for Real-Time Continual Learningcited by 97
  28. 2019Deep Learning for Audio Signal Processingcited by 92
  29. 2020Learning to Continually Learncited by 84
  30. 2018Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RLcited by 81
  31. 2019And the Bit Goes Down: Revisiting the Quantization of Neural Networkscited by 78
  32. 2021ActNN: Reducing Training Memory Footprint via 2-Bit Activation Compressed Trainingcited by 61
  33. 2019Low-Memory Neural Network Training: A Technical Reportcited by 57
  34. 2021Essentials for Class Incremental Learningcited by 57
  35. 2020μNAS: Constrained Neural Architecture Search for Microcontrollerscited by 55
  36. 2019AutoCompress: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Ratescited by 49
  37. 2022Enable Deep Learning on Mobile Devices: Methods, Systems, and Applicationscited by 32
  38. 2021RF-Net: a Unified Meta-learning Framework for RF-enabled One-shot Human Activity Recognitioncited by 24

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 4; 2025: 4; 2026: 3.

11 papers.

  1. 2024Self-Learning for Personalized Keyword Spotting on Ultra-Low-Power Audio Sensorscited by 7
  2. 2024Delta: A Cloud-assisted Data Enrichment Framework for On-Device Continual Learningcited by 6
  3. 2024Lifelong Intelligence Beyond the Edge using Hyperdimensional Computingcited by 5
  4. 2024Training on the Fly: On-device Self-supervised Learning aboard Nano-drones within 20 mWcited by 4
  5. 2025Ferret: An Efficient Online Continual Learning Framework under Varying Memory Constraintscited by 4
  6. 2025Energy-Aware Deep Learning on Resource-Constrained Hardwarecited by 3
  7. 2026ImageHD: Energy-Efficient On-Device Continual Learning of Visual Representations via Hyperdimensional Computingcited by 1
  8. 2025MetaCLBench: Meta Continual Learning Benchmark on Resource-Constrained Edge Devicescited by 0
  9. 2025M2RU: Memristive Minion Recurrent Unit for On-Chip Continual Learning at the Edgecited by 0
  10. 2026What changes after deployment? A survey on On-device Learning in TinyMLcited by 0
  11. 2026MicroQonv: Reshaping Convolution Tensors for Efficient Microscaling in Training and Inferencecited by 0