Eau De $Q$-Network: Adaptive Distillation of Neural Networks in Deep Reinforcement Learning

• URL: http://arxiv.org/abs/2503.01437v1
• Date: Mon, 03 Mar 2025 11:39:03 GMT
• Title: Eau De $Q$-Network: Adaptive Distillation of Neural Networks in Deep Reinforcement Learning
• Authors: Théo Vincent, Tim Faust, Yogesh Tripathi, Jan Peters, Carlo D'Eramo,
• Abstract summary: We propose a dense-to-sparse algorithm that increases sparsity at the agent's learning pace.<n>We evaluate the proposed approach on the Atari $2600$ benchmark and the MuJoCo physics simulator.
• Score: 17.356042621424567
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Recent works have successfully demonstrated that sparse deep reinforcement learning agents can be competitive against their dense counterparts. This opens up opportunities for reinforcement learning applications in fields where inference time and memory requirements are cost-sensitive or limited by hardware. Until now, dense-to-sparse methods have relied on hand-designed sparsity schedules that are not synchronized with the agent's learning pace. Crucially, the final sparsity level is chosen as a hyperparameter, which requires careful tuning as setting it too high might lead to poor performances. In this work, we address these shortcomings by crafting a dense-to-sparse algorithm that we name Eau De $Q$-Network (EauDeQN). To increase sparsity at the agent's learning pace, we consider multiple online networks with different sparsity levels, where each online network is trained from a shared target network. At each target update, the online network with the smallest loss is chosen as the next target network, while the other networks are replaced by a pruned version of the chosen network. We evaluate the proposed approach on the Atari $2600$ benchmark and the MuJoCo physics simulator, showing that EauDeQN reaches high sparsity levels while keeping performances high.

Related papers

• Active search and coverage using point-cloud reinforcement learning [50.741409008225766]
  This paper presents an end-to-end deep reinforcement learning solution for target search and coverage. We show that deep hierarchical feature learning works for RL and that by using farthest point sampling (FPS) we can reduce the amount of points. We also show that multi-head attention for point-clouds helps to learn the agent faster but converges to the same outcome.
  arXiv  Detail & Related papers  (2023-12-18T18:16:30Z)
• Learning a Consensus Sub-Network with Polarization Regularization and One Pass Training [2.895034191799291]
  Pruning schemes create extra overhead either by iterative training and fine-tuning for static pruning or repeated computation of a dynamic pruning graph.<n>We propose a new parameter pruning strategy for learning a lighter-weight sub-network that minimizes the energy cost while maintaining comparable performance to the fully parameterised network on given downstream tasks.<n>Our results on CIFAR-10, CIFAR-100, and Tiny Imagenet suggest that our scheme can remove 50% of connections in deep networks with 1% reduction in classification accuracy.
  arXiv  Detail & Related papers  (2023-02-17T09:37:17Z)
• DenseShift: Towards Accurate and Efficient Low-Bit Power-of-Two Quantization [27.231327287238102]
  We propose the DenseShift network, which significantly improves the accuracy of Shift networks. Our experiments on various computer vision and speech tasks demonstrate that DenseShift outperforms existing low-bit multiplication-free networks.
  arXiv  Detail & Related papers  (2022-08-20T15:17:40Z)
• Faster Deep Reinforcement Learning with Slower Online Network [90.34900072689618]
  We endow two popular deep reinforcement learning algorithms, namely DQN and Rainbow, with updates that incentivize the online network to remain in the proximity of the target network. The resultant agents, called DQN Pro and Rainbow Pro, exhibit significant performance improvements over their original counterparts on the Atari benchmark.
  arXiv  Detail & Related papers  (2021-12-10T22:07:06Z)
• Semi-supervised Network Embedding with Differentiable Deep Quantisation [81.49184987430333]
  We develop d-SNEQ, a differentiable quantisation method for network embedding. d-SNEQ incorporates a rank loss to equip the learned quantisation codes with rich high-order information. It is able to substantially compress the size of trained embeddings, thus reducing storage footprint and accelerating retrieval speed.
  arXiv  Detail & Related papers  (2021-08-20T11:53:05Z)
• FreeTickets: Accurate, Robust and Efficient Deep Ensemble by Training with Dynamic Sparsity [74.58777701536668]
  We introduce the FreeTickets concept, which can boost the performance of sparse convolutional neural networks over their dense network equivalents by a large margin. We propose two novel efficient ensemble methods with dynamic sparsity, which yield in one shot many diverse and accurate tickets "for free" during the sparse training process.
  arXiv  Detail & Related papers  (2021-06-28T10:48:20Z)
• Layer Folding: Neural Network Depth Reduction using Activation Linearization [0.0]
  Modern devices exhibit a high level of parallelism, but real-time latency is still highly dependent on networks' depth. We propose a method that learns whether non-linear activations can be removed, allowing to fold consecutive linear layers into one. We apply our method to networks pre-trained on CIFAR-10 and CIFAR-100 and find that they can all be transformed into shallower forms that share a similar depth.
  arXiv  Detail & Related papers  (2021-06-17T08:22:46Z)
• All at Once Network Quantization via Collaborative Knowledge Transfer [56.95849086170461]
  We develop a novel collaborative knowledge transfer approach for efficiently training the all-at-once quantization network. Specifically, we propose an adaptive selection strategy to choose a high-precision enquoteteacher for transferring knowledge to the low-precision student. To effectively transfer knowledge, we develop a dynamic block swapping method by randomly replacing the blocks in the lower-precision student network with the corresponding blocks in the higher-precision teacher network.
  arXiv  Detail & Related papers  (2021-03-02T03:09:03Z)
• Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networks [78.47459801017959]
  Sparsity can reduce the memory footprint of regular networks to fit mobile devices. We describe approaches to remove and add elements of neural networks, different training strategies to achieve model sparsity, and mechanisms to exploit sparsity in practice.
  arXiv  Detail & Related papers  (2021-01-31T22:48:50Z)
• HALO: Learning to Prune Neural Networks with Shrinkage [5.283963846188862]
  Deep neural networks achieve state-of-the-art performance in a variety of tasks by extracting a rich set of features from unstructured data. Modern techniques for inducing sparsity and reducing model size are (1) network pruning, (2) training with a sparsity inducing penalty, and (3) training a binary mask jointly with the weights of the network. We present a novel penalty called Hierarchical Adaptive Lasso which learns to adaptively sparsify weights of a given network via trainable parameters.
  arXiv  Detail & Related papers  (2020-08-24T04:08:48Z)
• Network Pruning via Annealing and Direct Sparsity Control [4.976007156860966]
  We propose a novel efficient network pruning method that is suitable for both non-structured and structured channel-level pruning. Our proposed method tightens a sparsity constraint by gradually removing network parameters or filter channels based on a criterion and a schedule.
  arXiv  Detail & Related papers  (2020-02-11T10:51:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.