Temporal Task Diversity: Inductive Biases Under Non-Stationarity in Synthetic Sequence Modelling

• URL: http://arxiv.org/abs/2605.18281v1
• Date: Mon, 18 May 2026 12:12:16 GMT
• Title: Temporal Task Diversity: Inductive Biases Under Non-Stationarity in Synthetic Sequence Modelling
• Abstract summary: In this paper, we explore the effect of diversifying the task distribution across training time.<n>We find that such temporal diversity leads to an increased bias towards generalisation over memorisation.
• Score: 3.330179707700564
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Modern deep learning science often assumes that neural networks learn from a fixed data distribution. However, many practically important learning problems involve data distributions that change throughout training. How does such non-stationarity impact the inductive biases of deep learning towards models with different structural, generalisation, and safety properties? A fruitful testbed for studying inductive bias is in-context linear regression sequence modelling, where small transformers display strikingly different generalisation patterns depending on the diversity of the (fixed) training task distribution. In this paper, we explore the effect of diversifying the task distribution across training time, finding that such temporal diversity leads to an increased bias towards generalisation over memorisation.

Related papers

• The Interplay of Data Structure and Imbalance in the Learning Dynamics of Diffusion Models [8.547042933482645]
  We study class-dependent learning in score-based diffusion models.<n>We reveal the explicit hierarchy governing these dynamics.<n>We validate our theoretical predictions empirically using U-Net models trained on Fashion MNIST.
  arXiv  Detail & Related papers  (2026-05-07T14:44:53Z)
• In-Context Linear Regression Demystified: Training Dynamics and Mechanistic Interpretability of Multi-Head Softmax Attention [52.159541540613915]
  We study how multi-head softmax attention models are trained to perform in-context learning on linear data.<n>Our results reveal that in-context learning ability emerges from the trained transformer as an aggregated effect of its architecture and the underlying data distribution.
  arXiv  Detail & Related papers  (2025-03-17T02:00:49Z)
• Generalizing to any diverse distribution: uniformity, gentle finetuning and rebalancing [55.791818510796645]
  We aim to develop models that generalize well to any diverse test distribution, even if the latter deviates significantly from the training data. Various approaches like domain adaptation, domain generalization, and robust optimization attempt to address the out-of-distribution challenge. We adopt a more conservative perspective by accounting for the worst-case error across all sufficiently diverse test distributions within a known domain.
  arXiv  Detail & Related papers  (2024-10-08T12:26:48Z)
• 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)
• Bias in Motion: Theoretical Insights into the Dynamics of Bias in SGD Training [7.5041863920639456]
  Machine learning systems often acquire biases by leveraging undesired features in the data, impacting accuracy across different sub-populations.<n>This paper explores the evolution of bias in a teacher-student setup modeling different data sub-populations with a Gaussian-mixture model.<n>Applying our findings to fairness and robustness, we delineate how and when heterogeneous data and spurious features can generate and amplify bias.
  arXiv  Detail & Related papers  (2024-05-28T15:50:10Z)
• DIVERSIFY: A General Framework for Time Series Out-of-distribution Detection and Generalization [58.704753031608625]
  Time series is one of the most challenging modalities in machine learning research. OOD detection and generalization on time series tend to suffer due to its non-stationary property. We propose DIVERSIFY, a framework for OOD detection and generalization on dynamic distributions of time series.
  arXiv  Detail & Related papers  (2023-08-04T12:27:11Z)
• Even Small Correlation and Diversity Shifts Pose Dataset-Bias Issues [19.4921353136871]
  We study two types of distribution shifts: diversity shifts, which occur when test samples exhibit patterns unseen during training, and correlation shifts, which occur when test data present a different correlation between seen invariant and spurious features. We propose an integrated protocol to analyze both types of shifts using datasets where they co-exist in a controllable manner.
  arXiv  Detail & Related papers  (2023-05-09T23:40:23Z)
• Generalized Representations Learning for Time Series Classification [28.230863650758447]
  We argue that the temporal complexity attributes to the unknown latent distributions within time series classification. We present experiments on gesture recognition, speech commands recognition, wearable stress and affect detection, and sensor-based human activity recognition.
  arXiv  Detail & Related papers  (2022-09-15T03:36:31Z)
• Distinguishing rule- and exemplar-based generalization in learning systems [10.396761067379195]
  We investigate two distinct inductive biases: feature-level bias and exemplar-vs-rule bias. We find that most standard neural network models have a propensity towards exemplar-based extrapolation. We discuss the implications of these findings for research on data augmentation, fairness, and systematic generalization.
  arXiv  Detail & Related papers  (2021-10-08T18:37:59Z)
• What training reveals about neural network complexity [80.87515604428346]
  This work explores the hypothesis that the complexity of the function a deep neural network (NN) is learning can be deduced by how fast its weights change during training. Our results support the hypothesis that good training behavior can be a useful bias towards good generalization.
  arXiv  Detail & Related papers  (2021-06-08T08:58:00Z)
• Learning from Failure: Training Debiased Classifier from Biased Classifier [76.52804102765931]
  We show that neural networks learn to rely on spurious correlation only when it is "easier" to learn than the desired knowledge. We propose a failure-based debiasing scheme by training a pair of neural networks simultaneously. Our method significantly improves the training of the network against various types of biases in both synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2020-07-06T07:20:29Z)

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.