Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userGST: Group-Sparse Training for Accelerating Deep Reinforcement Learning
98
0
0.0
(
0
)
Added by
Juhyoung Lee
Publication date
2021
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Deep reinforcement learning (DRL) has shown remarkable success in sequential decision-making problems but suffers from a long training time to obtain such good performance. Many parallel and distributed DRL training approaches have been proposed to solve this problem, but it is difficult to utilize them on resource-limited devices. In order to accelerate DRL in real-world edge devices, memory bandwidth bottlenecks due to large weight transactions have to be resolved. However, previous iterative pruning not only shows a low compression ratio at the beginning of training but also makes DRL training unstable. To overcome these shortcomings, we propose a novel weight compression method for DRL training acceleration, named group-sparse training (GST). GST selectively utilizes block-circulant compression to maintain a high weight compression ratio during all iterations of DRL training and dynamically adapt target sparsity through reward-aware pruning for stable training. Thanks to the features, GST achieves a 25 %p $sim$ 41.5 %p higher average compression ratio than the iterative pruning method without reward drop in Mujoco Halfcheetah-v2 and Mujoco humanoid-v2 environment with TD3 training.
rate research
Read More
Deep reinforcement learning has achieved significant success in many decision-making tasks in various fields. However, it requires a large training time of dense neural networks to obtain a good performance. This hinders its applicability on low-resource devices where memory and computation are strictly constrained. In a step towards enabling deep reinforcement learning agents to be applied to low-resource devices, in this work, we propose for the first time to dynamically train deep reinforcement learning agents with sparse neural networks from scratch. We adopt the evolution principles of dynamic sparse training in the reinforcement learning paradigm and introduce a training algorithm that optimizes the sparse topology and the weight values jointly to dynamically fit the incoming data. Our approach is easy to be integrated into existing deep reinforcement learning algorithms and has many favorable advantages. First, it allows for significant compression of the network size which reduces the memory and computation costs substantially. This would accelerate not only the agent inference but also its training process. Second, it speeds up the agent learning process and allows for reducing the number of required training steps. Third, it can achieve higher performance than training the dense counterpart network. We evaluate our approach on OpenAI gym continuous control tasks. The experimental results show the effectiveness of our approach in achieving higher performance than one of the state-of-art baselines with a 50% reduction in the network size and floating-point operations (FLOPs). Moreover, our proposed approach can reach the same performance achieved by the dense network with a 40-50% reduction in the number of training steps.
The success of deep learning in the computer vision and natural language processing communities can be attributed to training of very deep neural networks with millions or billions of parameters which can then be trained with massive amounts of data. However, similar trend has largely eluded training of deep reinforcement learning (RL) algorithms where larger networks do not lead to performance improvement. Previous work has shown that this is mostly due to instability during training of deep RL agents when using larger networks. In this paper, we make an attempt to understand and address training of larger networks for deep RL. We first show that naively increasing network capacity does not improve performance. Then, we propose a novel method that consists of 1) wider networks with DenseNet connection, 2) decoupling representation learning from training of RL, 3) a distributed training method to mitigate overfitting problems. Using this three-fold technique, we show that we can train very large networks that result in significant performance gains. We present several ablation studies to demonstrate the efficacy of the proposed method and some intuitive understanding of the reasons for performance gain. We show that our proposed method outperforms other baseline algorithms on several challenging locomotion tasks.
As neural network model sizes have dramatically increased, so has the interest in various techniques to reduce their parameter counts and accelerate their execution. An active area of research in this field is sparsity - encouraging zero values in parameters that can then be discarded from storage or computations. While most research focuses on high levels of sparsity, there are challenges in universally maintaining model accuracy as well as achieving significant speedups over modern matrix-math hardware. To make sparsity adoption practical, the NVIDIA Ampere GPU architecture introduces sparsity support in its matrix-math units, Tensor Cores. We present the design and behavior of Sparse Tensor Cores, which exploit a 2:4 (50%) sparsity pattern that leads to twice the math throughput of dense matrix units. We also describe a simple workflow for training networks that both satisfy 2:4 sparsity pattern requirements and maintain accuracy, verifying it on a wide range of common tasks and model architectures. This workflow makes it easy to prepare accurate models for efficient deployment on Sparse Tensor Cores.
We develop a parameterized Primal-Dual $pi$ Learning method based on deep neural networks for Markov decision process with large state space and off-policy reinforcement learning. In contrast to the popular Q-learning and actor-critic methods that are based on successive approximations to the nonlinear Bellman equation, our method makes primal-dual updates to the policy and value functions utilizing the fundamental linear Bellman duality. Naive parametrization of the primal-dual $pi$ learning method using deep neural networks would encounter two major challenges: (1) each update requires computing a probability distribution over the state space and is intractable; (2) the iterates are unstable since the parameterized Lagrangian function is no longer linear. We address these challenges by proposing a relaxed Lagrangian formulation with a regularization penalty using the advantage function. We show that the dual policy update step in our method is equivalent to the policy gradient update in the actor-critic method in some special case, while the value updates differ substantially. The main advantage of the primal-dual $pi$ learning method lies in that the value and policy updates are closely coupled together using the Bellman duality and therefore more informative. Experiments on a simple cart-pole problem show that the algorithm significantly outperforms the one-step temporal-difference actor-critic method, which is the most relevant benchmark method to compare with. We believe that the primal-dual updates to the value and policy functions would expedite the learning process. The proposed methods might open a door to more efficient algorithms and sharper theoretical analysis.
We investigate the hardness of online reinforcement learning in fixed horizon, sparse linear Markov decision process (MDP), with a special focus on the high-dimensional regime where the ambient dimension is larger than the number of episodes. Our contribution is two-fold. First, we provide a lower bound showing that linear regret is generally unavoidable in this case, even if there exists a policy that collects well-conditioned data. The lower bound construction uses an MDP with a fixed number of states while the number of actions scales with the ambient dimension. Note that when the horizon is fixed to one, the case of linear stochastic bandits, the linear regret can be avoided. Second, we show that if the learner has oracle access to a policy that collects well-conditioned data then a variant of Lasso fitted Q-iteration enjoys a nearly dimension-free regret of $tilde{O}( s^{2/3} N^{2/3})$ where $N$ is the number of episodes and $s$ is the sparsity level. This shows that in the large-action setting, the difficulty of learning can be attributed to the difficulty of finding a good exploratory policy.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
