Functional Regularization For Reinforcement Learning Via Learned Fourier Features

Abstract

We propose a simple architecture for deep reinforcement learning by embedding inputs into a learned Fourier basis and show that it improves the sample efficiency of both state-based and image-based RL. We perform infinite-width analysis of our architecture using the Neural Tangent Kernel and theoretically show that tuning the initial variance of the Fourier basis is equivalent to functional regularization of the learned deep network. That is, these learned Fourier features allow for adjusting the degree to which networks underfit or overfit different frequencies in the training data, and hence provide a controlled mechanism to improve the stability and performance of RL optimization. Empirically, this allows us to prioritize learning low-frequency functions and speed up learning by reducing networks' susceptibility to noise in the optimization process, such as during Bellman updates. Experiments on standard state-based and image-based RL benchmarks show clear benefits of our architectu

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyli2021functional

Related papers

• Frequency And Generalisation Of Periodic Activation Functions In Reinforcement Learning (2024)0.00
• Regularization Matters In Policy Optimization (2019)2.68
• Spectral Representation-based Reinforcement Learning (2025)0.00
• State Sequences Prediction Via Fourier Transform For Representation Learning (2023)2.26
• Spectral Normalisation For Deep Reinforcement Learning: An Optimisation Perspective (2021)0.00
• Reinforcement Learning In Feature Space: Matrix Bandit, Kernels, And Regret Bound (2019)0.00
• Temporal Regularization In Markov Decision Process (2018)0.00
• Adapting The Function Approximation Architecture In Online Reinforcement Learning (2021)0.00