Learning to Focus: Prioritizing Informative Histories with Structured Attention Mechanisms in Partially Observable Reinforcement Learning

• URL: http://arxiv.org/abs/2511.06946v1
• Date: Mon, 10 Nov 2025 10:53:16 GMT
• Title: Learning to Focus: Prioritizing Informative Histories with Structured Attention Mechanisms in Partially Observable Reinforcement Learning
• Authors: Daniel De Dios Allegue, Jinke He, Frans A. Oliehoek,
• Abstract summary: We introduce structured inductive priors into the self-attention mechanism of the dynamics head.<n>Experiments on the Atari 100k benchmark show that most efficiency gains arise from the Gaussian prior.
• Score: 9.233407096706744
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Transformers have shown strong ability to model long-term dependencies and are increasingly adopted as world models in model-based reinforcement learning (RL) under partial observability. However, unlike natural language corpora, RL trajectories are sparse and reward-driven, making standard self-attention inefficient because it distributes weight uniformly across all past tokens rather than emphasizing the few transitions critical for control. To address this, we introduce structured inductive priors into the self-attention mechanism of the dynamics head: (i) per-head memory-length priors that constrain attention to task-specific windows, and (ii) distributional priors that learn smooth Gaussian weightings over past state-action pairs. We integrate these mechanisms into UniZero, a model-based RL agent with a Transformer-based world model that supports planning under partial observability. Experiments on the Atari 100k benchmark show that most efficiency gains arise from the Gaussian prior, which smoothly allocates attention to informative transitions, while memory-length priors often truncate useful signals with overly restrictive cut-offs. In particular, Gaussian Attention achieves a 77% relative improvement in mean human-normalized scores over UniZero. These findings suggest that in partially observable RL domains with non-stationary temporal dependencies, discrete memory windows are difficult to learn reliably, whereas smooth distributional priors flexibly adapt across horizons and yield more robust data efficiency. Overall, our results demonstrate that encoding structured temporal priors directly into self-attention improves the prioritization of informative histories for dynamics modeling under partial observability.

Related papers

• VARMA-Enhanced Transformer for Time Series Forecasting [4.982130518684668]
  VARMAformer is a novel architecture that synergizes the efficiency of a cross-attention-only framework with the principles of classical time series analysis.<n>By fusing these classical insights into a modern backbone, VARMAformer captures both global, long-range dependencies and local, statistical structures.
  arXiv  Detail & Related papers  (2025-09-05T03:32:51Z)
• SelfAug: Mitigating Catastrophic Forgetting in Retrieval-Augmented Generation via Distribution Self-Alignment [49.86376148975563]
  Large language models (LLMs) have revolutionized natural language processing through their capabilities in understanding and executing diverse tasks.<n> supervised fine-tuning, particularly in Retrieval-Augmented Generation (RAG) scenarios, often leads to catastrophic forgetting.<n>We propose SelfAug, a self-distribution alignment method that aligns input sequence logits to preserve the model's semantic distribution.
  arXiv  Detail & Related papers  (2025-09-04T06:50:47Z)
• Why Generate When You Can Transform? Unleashing Generative Attention for Dynamic Recommendation [9.365893765448366]
  Sequential Recommendation (SR) focuses on personalizing user experiences by predicting future preferences based on historical interactions.<n> Transformer models, with their attention mechanisms, have become the dominant architecture in SR tasks.<n>We introduce two generative attention models for SR, each grounded in the principles of Variational Autoencoders (VAE) and Diffusion Models (DMs)
  arXiv  Detail & Related papers  (2025-08-04T04:33:26Z)
• Detecting and Pruning Prominent but Detrimental Neurons in Large Language Models [68.57424628540907]
  Large language models (LLMs) often develop learned mechanisms specialized to specific datasets.<n>We introduce a fine-tuning approach designed to enhance generalization by identifying and pruning neurons associated with dataset-specific mechanisms.<n>Our method employs Integrated Gradients to quantify each neuron's influence on high-confidence predictions, pinpointing those that disproportionately contribute to dataset-specific performance.
  arXiv  Detail & Related papers  (2025-07-12T08:10:10Z)
• Powerformer: A Transformer with Weighted Causal Attention for Time-series Forecasting [50.298817606660826]
  We introduce Powerformer, a novel Transformer variant that replaces noncausal attention weights with causal weights that are reweighted according to a smooth heavy-tailed decay.<n>Our empirical results demonstrate that Powerformer achieves state-of-the-art accuracy on public time-series benchmarks.<n>Our analyses show that the model's locality bias is amplified during training, demonstrating an interplay between time-series data and power-law-based attention.
  arXiv  Detail & Related papers  (2025-02-10T04:42:11Z)
• RecurFormer: Not All Transformer Heads Need Self-Attention [14.331807060659902]
  Transformer-based large language models (LLMs) excel in modeling complex language patterns but face significant computational costs during inference. We propose RecurFormer, a novel architecture that replaces certain attention heads with linear recurrent neural networks.
  arXiv  Detail & Related papers  (2024-10-10T15:24:12Z)
• WAVE: Weighted Autoregressive Varying Gate for Time Series Forecasting [9.114664059026767]
  We propose a weighted Autoregressive Varying gatE attention mechanism equipped with both Autoregressive (AR) and Moving-average (MA) components.<n>It can adapt to various attention mechanisms, enhancing and decoupling their ability to capture long-range and local temporal patterns in time series data.
  arXiv  Detail & Related papers  (2024-10-04T05:45:50Z)
• Enhancing Robustness of Vision-Language Models through Orthogonality Learning and Self-Regularization [77.62516752323207]
  We introduce an orthogonal fine-tuning method for efficiently fine-tuning pretrained weights and enabling enhanced robustness and generalization. A self-regularization strategy is further exploited to maintain the stability in terms of zero-shot generalization of VLMs, dubbed OrthSR. For the first time, we revisit the CLIP and CoOp with our method to effectively improve the model on few-shot image classficiation scenario.
  arXiv  Detail & Related papers  (2024-07-11T10:35:53Z)
• Attention as Robust Representation for Time Series Forecasting [23.292260325891032]
  Time series forecasting is essential for many practical applications. Transformers' key feature, the attention mechanism, dynamically fusing embeddings to enhance data representation, often relegating attention weights to a byproduct role. Our approach elevates attention weights as the primary representation for time series, capitalizing on the temporal relationships among data points to improve forecasting accuracy.
  arXiv  Detail & Related papers  (2024-02-08T03:00:50Z)
• TransNormerLLM: A Faster and Better Large Language Model with Improved TransNormer [34.790081960470964]
  We present TransNormerLLM, the first linear attention-based Large Language Model (LLM) We make advanced modifications that include positional embedding, linear attention acceleration, gating mechanisms, tensor normalization, and inference acceleration and stabilization. We validate our model design through a series of ablations and train models with sizes of 385M, 1B, and 7B on our self-collected corpus.
  arXiv  Detail & Related papers  (2023-07-27T16:45:33Z)
• INFOrmation Prioritization through EmPOWERment in Visual Model-Based RL [90.06845886194235]
  We propose a modified objective for model-based reinforcement learning (RL) We integrate a term inspired by variational empowerment into a state-space model based on mutual information. We evaluate the approach on a suite of vision-based robot control tasks with natural video backgrounds.
  arXiv  Detail & Related papers  (2022-04-18T23:09:23Z)
• Cauchy-Schwarz Regularized Autoencoder [68.80569889599434]
  Variational autoencoders (VAE) are a powerful and widely-used class of generative models. We introduce a new constrained objective based on the Cauchy-Schwarz divergence, which can be computed analytically for GMMs. Our objective improves upon variational auto-encoding models in density estimation, unsupervised clustering, semi-supervised learning, and face analysis.
  arXiv  Detail & Related papers  (2021-01-06T17:36:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.