paper-with-me

홈 › Papers

An Alternate Policy Gradient Estimator for Softmax Policies

2021-12-22 · Shivam Garg, Samuele Tosatto, Yangchen Pan, Martha White, A. Rupam Mahmood

Policy gradient (PG) estimators are ineffective in dealing with softmax policies that are sub-optimally saturated, which refers to the situation when the policy concentrates its probability mass on sub-optimal actions. Sub-optimal policy saturation may arise from bad policy initialization or sudden changes in the environment that occur after the policy has already converged. Current softmax PG estimators require a large number of updates to overcome policy saturation, which causes low sample efficiency and poor adaptability to new situations. To mitigate this problem, we propose a novel PG estimator for softmax policies that utilizes the bias in the critic estimate and the noise present in the reward signal to escape the saturated regions of the policy parameter space. Our theoretical analysis and experiments, conducted on bandits and various reinforcement learning environments, show that this new estimator is significantly more robust to policy saturation.

📄 PDF Abstract BibTeX arXiv:2112.11622

Code (1)

svmgrg/alternate_pg 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

Differentiable Meta-Learning of Bandit Policies

2020-12-01 · NeurIPS 2020 12 · Craig Boutilier, Chih-Wei Hsu, Branislav Kveton, Martin Mladenov 외

Exploration policies in Bayesian bandits maximize the average reward over problem instances drawn from some distribution P. In this work, we learn such policies for an unknown distribution P using samples from P. Our app…

Meta-Learning

Differentiable Bandit Exploration

2020-02-17 · NeurIPS 2020 12 · Craig Boutilier, Chih-Wei Hsu, Branislav Kveton, Martin Mladenov 외

Exploration policies in Bayesian bandits maximize the average reward over problem instances drawn from some distribution $\mathcal{P}$. In this work, we learn such policies for an unknown distribution $\mathcal{P}$ using…

Meta-Learning

Convergence and Price of Anarchy Guarantees of the Softmax Policy Gradient in Markov Potential Games

2022-06-15 · Dingyang Chen, Qi Zhang, Thinh T. Doan

We study the performance of policy gradient methods for the subclass of Markov games known as Markov potential games (MPGs), which extends the notion of normal-form potential games to the stateful setting and includes th…

Policy Gradient Methods

Softmax Deep Double Deterministic Policy Gradients

2020-10-19 · NeurIPS 2020 12 · Ling Pan, Qingpeng Cai, Longbo Huang

A widely-used actor-critic reinforcement learning algorithm for continuous control, Deep Deterministic Policy Gradients (DDPG), suffers from the overestimation problem, which can negatively affect the performance. Althou…

continuous-controlContinuous Control

Beyond Stationarity: Convergence Analysis of Stochastic Softmax Policy Gradient Methods

2023-10-04 · Sara Klein, Simon Weissmann, Leif Döring

Markov Decision Processes (MDPs) are a formal framework for modeling and solving sequential decision-making problems. In finite-time horizons such problems are relevant for instance for optimal stopping or specific suppl…

Decision MakingPolicy Gradient MethodsSequential Decision Making