paper-with-me

Papers

Implicit Under-Parameterization Inhibits Data-Efficient Deep Reinforcement Learning

2020-10-27 · ICLR 2021 1 · Aviral Kumar, Rishabh Agarwal, Dibya Ghosh, Sergey Levine

We identify an implicit under-parameterization phenomenon in value-based deep RL methods that use bootstrapping: when value functions, approximated using deep neural networks, are trained with gradient descent using iterated regression onto target values generated by previous instances of the value network, more gradient updates decrease the expressivity of the current value network. We characterize this loss of expressivity via a drop in the rank of the learned value network features, and show that this typically corresponds to a performance drop. We demonstrate this phenomenon on Atari and Gym benchmarks, in both offline and online RL settings. We formally analyze this phenomenon and show that it results from a pathological interaction between bootstrapping and gradient-based optimization. We further show that mitigating implicit under-parameterization by controlling rank collapse can improve performance.

📄 PDF Abstract BibTeX arXiv:2010.14498

Code (1)

google-research/batch_rl tf

Tasks

Deep Reinforcement Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Similar Papers 제목 키워드 기반

Reward-Conditioned Reinforcement Learning

2026-03-05 · Michal Nauman, Marek Cygan, Pieter Abbeel arxiv

Single-task RL agents are typically trained under a fixed reward function, which limits their robustness to reward misspecification and their ability to adapt to changing preferences. We introduce Reward-Conditioned Rein…

Reinforcement LearningMulti-Task Learning

Swapout: Learning an ensemble of deep architectures

2016-05-20 · NeurIPS 2016 12 · Saurabh Singh, Derek Hoiem, David Forsyth

We describe Swapout, a new stochastic training method, that outperforms ResNets of identical network structure yielding impressive results on CIFAR-10 and CIFAR-100. Swapout samples from a rich set of architectures inclu…

Soft Actor-Critic with Beta Policy via Implicit Reparameterization Gradients

2024-09-08 · Luca Della Libera

Recent advances in deep reinforcement learning have achieved impressive results in a wide range of complex tasks, but poor sample efficiency remains a major obstacle to real-world deployment. Soft actor-critic (SAC) miti…

continuous-controlContinuous ControlDeep Reinforcement Learning

Dissipative residual layers for unsupervised implicit parameterization of data manifolds

2022-10-13 · Viktor Reshniak

We propose an unsupervised technique for implicit parameterization of data manifolds. In our approach, the data is assumed to belong to a lower dimensional manifold in a higher dimensional space, and the data points are …

DenoisingReinforcement Learning (RL)

Explicit Preference Optimization: No Need for an Implicit Reward Model

2025-06-09 · Xiangkun Hu, Lemin Kong, Tong He, David Wipf

The generated responses of large language models (LLMs) are often fine-tuned to human preferences through a process called reinforcement learning from human feedback (RLHF). As RLHF relies on a challenging training seque…