mlsanityData to 14 Oct 2026

A Method of Selective Attention for Reservoir Based Agents

arXiv 2502.21229 ↗First posted 28 Feb 2025cs.LGcited by 1

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations

Abstract (arXiv)

Training of deep reinforcement learning agents is slowed considerably by the presence of input dimensions that do not usefully condition the reward function. Existing modules such as layer normalization can be trained with weight decay to act as a form of selective attention, i.e. an input mask, that shrinks the scale of unnecessary inputs, which in turn accelerates training of the policy. However, we find a surprising result that adding numerous parameters to the computation of the input mask results in much faster training. A simple, high dimensional masking module is compared with layer normalization and a model without any input suppression. The high dimensional mask resulted in a four-fold speedup in training over the null hypothesis and a two-fold speedup in training over the layer normalization method.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 2 of its 8 references that are in the map. First posted on arXiv 28 Feb 2025.

Builds on

2 of 8 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Reservoir Computing for Fast, Simplified Reinforcement Learning on Memory Tasks17 Dec

Other references · 1

  1. 2016Layer Normalizationcited by 5,669

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Flexible Prefrontal Control over Hippocampal Episodic Memory for Goal-Directed Generalizationcited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2026Sparse Masked Attention Policies for Reliable Generalization
  2. 2026Mask-based Predictive Representations for Reinforcement Learning
  3. 2025Optimizing Sensory Neurons: Nonlinear Attention Mechanisms for Accelerated Convergence in Permutation-Invariant Neural Networks for Reinforcement Learning
  4. 2025Deep Reinforcement Learning via Object-Centric Attention
  5. 2021Hard Attention Control By Mutual Information Maximization
  6. 2024Excluding the Irrelevant: Focusing Reinforcement Learning through Continuous Action Masking
  7. 2022Masked Autoencoding for Scalable and Generalizable Decision Making
  8. 2025Experience-Efficient Model-Free Deep Reinforcement Learning Using Pre-Training