Understanding Generalization through Decision Pattern Shift

• URL: http://arxiv.org/abs/2605.13148v1
• Date: Wed, 13 May 2026 08:14:59 GMT
• Title: Understanding Generalization through Decision Pattern Shift
• Abstract summary: Decision Pattern Shift (DPS) is a new perspective that defines generalization through the stability of internal decision patterns.<n>We show that DPS quantifies failure as their deviation from those learned during training.<n>These findings open up new possibilities for early generalization-risk detection, failure-mode diagnosis, and channel-level defect localization.
• Score: 46.36674063889302
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding why deep neural networks (DNNs) fail to generalize to unseen samples remains a long-standing challenge. Existing studies mainly examine changes in externally observable factors such as data, representations, or outputs, yet offer limited insight into how a model's internal decision mechanism evolves from training to test. To address this gap, we introduce Decision Pattern Shift (DPS), a new perspective that defines generalization through the stability of internal decision patterns and quantifies failure as their deviation from those learned during training. Specifically, we represent each sample's decision pattern as a GradCAM-based channel-contribution vector, which captures how feature channels collectively support a prediction, and we propose the DPS metric to measure its discrepancy from the class-average pattern. Empirical analyses across multiple datasets and architectures show that, (i) decision patterns form a highly structured, class-consistent space with strong intra-class cohesion and low inter-class confusion, enabling direct analysis of a model's decision logic; (ii) the DPS magnitude correlates linearly with the generalization gap (nearly all Pearson r > 0.8), revealing generalization as a systematic drift in the model's internal decision mechanism; (iii) the DPS spectrum organizes diverse generalization degradation scenarios (covering ideal generalization, in-distribution degradation, domain shift, out-of-distribution, and shortcut learning) into a continuous trajectory, providing a unified explanation of their failure modes. These findings open up new possibilities for early generalization-risk detection, failure-mode diagnosis, and channel-level defect localization.

Related papers

• Learning to Trim: End-to-End Causal Graph Pruning with Dynamic Anatomical Feature Banks for Medical VQA [27.877350391080693]
  We propose a Learnable Causal Trimming (LCT) framework that integrates causal pruning into end-to-end optimization.<n>We introduce a Dynamic Anatomical Feature Bank (DAFB), updated via a momentum mechanism, to capture global prototypes of frequent anatomical and linguistic patterns.<n>Experiments on VQA-RAD, SLAKE, SLAKE-CP and PathVQA demonstrate that LCT consistently improves robustness and generalization over existing debiasing strategies.
  arXiv  Detail & Related papers  (2026-03-27T02:57:39Z)
• Generalization Analysis and Method for Domain Generalization for a Family of Recurrent Neural Networks [11.316438734182931]
  This paper proposes a method to analyze interpretability and out-of-domain (OOD) generalization for a family of recurrent neural networks (RNNs)<n>Specifically, the evolution of a trained RNN's states is modeled as an unknown, discrete-time, nonlinear closed-loop feedback system.<n>A domain generalization method is proposed that reduces the OOD generalization error and improves the robustness to distribution shifts.
  arXiv  Detail & Related papers  (2026-01-13T01:27:10Z)
• Learning Time-Aware Causal Representation for Model Generalization in Evolving Domains [50.66049136093248]
  We develop a time-aware structural causal model (SCM) that incorporates dynamic causal factors and the causal mechanism drifts.<n>We show that our method can yield the optimal causal predictor for each time domain.<n>Results on both synthetic and real-world datasets exhibit that SYNC can achieve superior temporal generalization performance.
  arXiv  Detail & Related papers  (2025-06-21T14:05:37Z)
• DeCaf: A Causal Decoupling Framework for OOD Generalization on Node Classification [14.96980804513399]
  Graph Neural Networks (GNNs) are susceptible to distribution shifts, creating vulnerability and security issues in critical domains. Existing methods that target learning an invariant (feature, structure)-label mapping often depend on oversimplified assumptions about the data generation process. We introduce a more realistic graph data generation model using Structural Causal Models (SCMs) We propose a casual decoupling framework, DeCaf, that independently learns unbiased feature-label and structure-label mappings.
  arXiv  Detail & Related papers  (2024-10-27T00:22:18Z)
• Learning Divergence Fields for Shift-Robust Graph Representations [73.11818515795761]
  In this work, we propose a geometric diffusion model with learnable divergence fields for the challenging problem with interdependent data. We derive a new learning objective through causal inference, which can guide the model to learn generalizable patterns of interdependence that are insensitive across domains.
  arXiv  Detail & Related papers  (2024-06-07T14:29:21Z)
• GEC: A Unified Framework for Interactive Decision Making in MDP, POMDP, and Beyond [101.5329678997916]
  We study sample efficient reinforcement learning (RL) under the general framework of interactive decision making. We propose a novel complexity measure, generalized eluder coefficient (GEC), which characterizes the fundamental tradeoff between exploration and exploitation. We show that RL problems with low GEC form a remarkably rich class, which subsumes low Bellman eluder dimension problems, bilinear class, low witness rank problems, PO-bilinear class, and generalized regular PSR.
  arXiv  Detail & Related papers  (2022-11-03T16:42:40Z)
• On the Generalization and Adaption Performance of Causal Models [99.64022680811281]
  Differentiable causal discovery has proposed to factorize the data generating process into a set of modules. We study the generalization and adaption performance of such modular neural causal models. Our analysis shows that the modular neural causal models outperform other models on both zero and few-shot adaptation in low data regimes.
  arXiv  Detail & Related papers  (2022-06-09T17:12:32Z)
• TsmoBN: Interventional Generalization for Unseen Clients in Federated Learning [23.519212374186232]
  We form a training structural causal model (SCM) to explain the challenges of model generalization in a distributed learning paradigm. We present a simple yet effective method using test-specific and momentum tracked batch normalization (TsmoBN) to generalize FL models to testing clients.
  arXiv  Detail & Related papers  (2021-10-19T13:46:37Z)
• Invariant Causal Prediction for Block MDPs [106.63346115341862]
  Generalization across environments is critical to the successful application of reinforcement learning algorithms to real-world challenges. We propose a method of invariant prediction to learn model-irrelevance state abstractions (MISA) that generalize to novel observations in the multi-environment setting.
  arXiv  Detail & Related papers  (2020-03-12T21:03:01Z)

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.