mlsanityData to 7 Oct 2026

p-Meta: Towards On-device Deep Model Adaptation

arXiv 2206.12705 ↗First posted 25 Jun 2022cs.LG cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Data collected by IoT devices are often private and have a large diversity across users. Therefore, learning requires pre-training a model with available representative data samples, deploying the pre-trained model on IoT devices, and adapting the deployed model on the device with local data. Such an on-device adaption for deep learning empowered applications demands data and memory efficiency. However, existing gradient-based meta learning schemes fail to support memory-efficient adaptation. To this end, we propose p-Meta, a new meta learning method that enforces structure-wise partial parameter updates while ensuring fast generalization to unseen tasks. Evaluations on few-shot image classification and reinforcement learning tasks show that p-Meta not only improves the accuracy but also substantially reduces the peak dynamic memory by a factor of 2.5 on average compared to state-of-the-art few-shot adaptation methods.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 29 of its 43 references that are in the map. First posted on arXiv 25 Jun 2022.

Builds on

29 of 43 references.

References · 29

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  4. 2017Squeeze-and-Excitation Networkscited by 4,763
  5. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  6. 2015Trust Region Policy Optimizationcited by 2,503
  7. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
  8. 2016Matching Networks for One Shot Learningcited by 2,246
  9. 2018Group Normalizationcited by 1,514
  10. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  11. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  12. 2020Meta-Learning in Neural Networks: A Surveycited by 944
17 more
  1. 2016Training Deep Nets with Sublinear Memory Costcited by 825
  2. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  3. 2018TADAM: Task dependent adaptive metric for improved few-shot learningcited by 408
  4. 2016Benchmarking Deep Reinforcement Learning for Continuous Controlcited by 361
  5. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  6. 2019Dynamic Convolution: Attention over Convolution Kernelscited by 355
  7. 2019Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examplescited by 312
  8. 2018How to train your MAMLcited by 306
  9. 2017The Reversible Residual Network: Backpropagation Without Storing Activationscited by 215
  10. 2016Memory-Efficient Backpropagation Through Timecited by 78
  11. 2021Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learningcited by 72
  12. 2016Highway and Residual Networks learn Unrolled Iterative Estimationcited by 52
  13. 2019Torchmeta: A Meta-Learning library for PyTorchcited by 52
  14. 2020Sparse Weight Activation Trainingcited by 45
  15. 2021Learning where to learn: Gradient sparsity in meta and continual learningcited by 33
  16. 2020Low-rank Gradient Approximation For Memory-Efficient On-device Training of Deep Neural Networkcited by 20
  17. 2021On-device Federated Learning with Flowercited by 13

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 3; 2024: 2; 2025: 2.

8 papers.

  1. 2022On-device Training: A First Overview on Existing Systemscited by 18
  2. 2023TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edgecited by 18
  3. 2023LifeLearner: Hardware-Aware Meta Continual Learning System for Embedded Computing Platformscited by 11
  4. 2024Training Machine Learning models at the Edge: A Surveycited by 11
  5. 2023Hierarchical Training of Deep Neural Networks Using Early Exitingcited by 5
  6. 2024AdaShadow: Responsive Test-time Model Adaptation in Non-stationary Mobile Environmentscited by 1
  7. 2025Tin-Tin: Towards Tiny Learning on Tiny Devices with Integer-based Neural Network Trainingcited by 1
  8. 2025Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layerscited by 0