mlsanityData to 14 Oct 2026

Neural Network Compression for Reinforcement Learning Tasks

arXiv 2405.07748 ↗First posted 13 May 2024cs.LGcited by 3

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations

Abstract (arXiv)

In real applications of Reinforcement Learning (RL), such as robotics, low latency and energy efficient inference is very desired. The use of sparsity and pruning for optimizing Neural Network inference, and particularly to improve energy and latency efficiency, is a standard technique. In this work, we perform a systematic investigation of applying these optimization techniques for different RL algorithms in different RL environments, yielding up to a 400-fold reduction in the size of neural networks.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 29 of its 61 references that are in the map. First posted on arXiv 13 May 2024.

Builds on

29 of 61 references.

References · 29

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  3. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  4. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  5. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  6. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 2,031
  7. 2018IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architecturescited by 883
  8. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 815
  9. 2018SNIP: Single-shot Network Pruning based on Connection Sensitivitycited by 812
  10. 2017To prune, or not to prune: exploring the efficacy of pruning for model compressioncited by 603
  11. 2015BinaryConnect: Training Deep Neural Networks with binary weights during propagationscited by 592
  12. 2020What is the State of Neural Network Pruning?cited by 583
17 more
  1. 2019Rigging the Lottery: Making All Tickets Winnerscited by 495
  2. 2020Picking Winning Tickets Before Training by Preserving Gradient Flowcited by 471
  3. 2017Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Sciencecited by 328
  4. 2016EIE: Efficient Inference Engine on Compressed Deep Neural Networkcited by 295
  5. 2017Variational Dropout Sparsifies Deep Neural Networkscited by 265
  6. 2021Pruning and Quantization for Deep Neural Network Acceleration: A Surveycited by 263
  7. 2020A Comprehensive Survey of Neural Architecture Search: Challenges and Solutionscited by 254
  8. 2019Understanding Straight-Through Estimator in Training Activation Quantized Neural Netscited by 231
  9. 2017Deep Rewiring: Training very sparse deep networkscited by 139
  10. 2017SCNN: An Accelerator for Compressed-sparse Convolutional Neural Networkscited by 135
  11. 2017Bayesian Compression for Deep Learningcited by 116
  12. 2018Faster gaze prediction with dense networks and Fisher pruningcited by 94
  13. 2022The State of Sparse Training in Deep Reinforcement Learningcited by 61
  14. 2018Compressing Neural Networks using the Variational Information Bottleneckcited by 59
  15. 2019QuaRL: Quantization for Fast and Environmentally Sustainable Reinforcement Learningcited by 19
  16. 2022Neuromorphic Artificial Intelligence Systemscited by 12
  17. 2018Is Information in the Brain Represented in Continuous or Discrete Form?cited by 4

Built on by

Cited by year

Citing papers per year: 2025: 3.

3 papers.

  1. 2025Learning Quantized Continuous Controllers for Integer Hardwarecited by 1
  2. 2025Differentiable Weightless Controllers: Learning Logic Circuits for Continuous Controlcited by 1
  3. 2025Emergence of hybrid computational dynamics through reinforcement learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2021Single-Shot Pruning for Offline Reinforcement Learning
  2. 2024The Impact of Quantization and Pruning on Deep Reinforcement Learning Models
  3. 2022Towards Hardware-Specific Automatic Compression of Neural Networks
  4. 2026RNM-TD3: N:M Semi-structured Sparse Reinforcement Learning From Scratch
  5. 2025Online Training and Pruning of Deep Reinforcement Learning Networks
  6. 2026Hierarchical Reinforcement Learning for Neural Network Compression (HiReLC): Pruning and Quantization
  7. 2025Network Sparsity Unlocks the Scaling Potential of Deep Reinforcement Learning
  8. 2024Optimal Policy Sparsification and Low Rank Decomposition for Deep Reinforcement Learning