Towards Out-Of-Distribution Generalization: A Survey
read the original abstract
Traditional machine learning paradigms are based on the assumption that both training and test data follow the same statistical pattern, which is mathematically referred to as Independent and Identically Distributed ($i.i.d.$). However, in real-world applications, this $i.i.d.$ assumption often fails to hold due to unforeseen distributional shifts, leading to considerable degradation in model performance upon deployment. This observed discrepancy indicates the significance of investigating the Out-of-Distribution (OOD) generalization problem. OOD generalization is an emerging topic of machine learning research that focuses on complex scenarios wherein the distributions of the test data differ from those of the training data. This paper represents the first comprehensive, systematic review of OOD generalization, encompassing a spectrum of aspects from problem definition, methodological development, and evaluation procedures, to the implications and future directions of the field. Our discussion begins with a precise, formal characterization of the OOD generalization problem. Following that, we categorize existing methodologies into three segments: unsupervised representation learning, supervised model learning, and optimization, according to their positions within the overarching learning process. We provide an in-depth discussion on representative methodologies for each category, further elucidating the theoretical links between them. Subsequently, we outline the prevailing benchmark datasets employed in OOD generalization studies. To conclude, we overview the existing body of work in this domain and suggest potential avenues for future research on OOD generalization. A summary of the OOD generalization methodologies surveyed in this paper can be accessed at http://out-of-distribution-generalization.com.
This paper has not been read by Pith yet.
Forward citations
Cited by 19 Pith papers
-
Discovery of unobservable parameters via physical embedding
PEIL learns unobservable parameters by embedding them in a physics-based reconstruction loop, outperforming supervised baselines with ground-truth access while enabling zero-shot generalization and major data reductio...
-
Scaling Nonlinear Optimization: Many Problems One GPU
jaxipm is the first GPU-batched IPOPT solver in JAX using heterogeneous iteration fusion and iteration-level batching, delivering up to 32.85x higher throughput than standard IPOPT on quadrotor NMPC benchmarks.
-
eXplaining to Learn (eX2L): Regularization Using Contrastive Visual Explanation Pairs for Distribution Shifts
eX2L improves robustness to distribution shifts by penalizing similarity between Grad-CAM maps of a label classifier and a confounder classifier, reaching new SOTA average and worst-group accuracy on the Spawrious benchmark.
-
UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models
UDM-GRPO is the first RL integration for uniform discrete diffusion models, using final clean samples as actions and forward-process trajectory reconstruction to raise GenEval accuracy from 69% to 96% and OCR accuracy...
-
OPD-Evolver: Cultivating Holistic Agent Evolver via On-Policy Distillation
OPD-Evolver uses on-policy self-distillation in fast interaction and slow attribution loops to build agents with holistic memory competence, outperforming prior systems by up to 11.5% and allowing a 9B model to compet...
-
Rethinking Molecular OOD Generalization via Target-Aware Source Selection
SCOPE-BENCH shows state-of-the-art molecular models suffer up to 8x higher errors under extreme OOD, while POMA reduces mean absolute error by up to 11.2% via target-aware source selection and dual-scale adaptation.
-
UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models
UDM-GRPO integrates reinforcement learning with Uniform Discrete Diffusion Models by treating the final clean sample as the action and reconstructing trajectories via the forward process, achieving state-of-the-art te...
-
Adversarial Label Invariant Graph Data Augmentations for Out-of-Distribution Generalization
RIA uses adversarial exploration of counterfactual graph environments via label-invariant augmentations to improve OoD generalization in graph classification tasks.
-
CausalMix: Data Mixture as Causal Inference for Language Model Training
CausalMix fits a causal model on 512 runs of a 0.5B model to estimate CATE, then extrapolates optimal mixtures for an 800K data pool applied to 7B and 4B models, outperforming RegMix.
-
Human-Machine Collaboration on Generative Meta-Learning: Model and Algorithm
GMHF aligns generated data distributions with human beliefs about target physics using cNODE and RL to reduce generalization risk, validated on Duffing oscillator and probabilistic models.
-
Physics-Informed Attention Mechanism and Generalization Capability of Deep Learning-Based Grain Growth Evolution Prediction
Deep learning models for grain growth generalize to out-of-distribution microstructures without retraining, with a boundary-masked attention mechanism improving SSIM and grain size error on bimodal cases.
-
CA-DEL: An Open Multi-Target, Multi-Modal Benchmark for Learning from DNA-Encoded Library Screens
CA-DEL is a new multi-target benchmark for training ML models on noisy DEL sequencing data and evaluating them on high-fidelity Ki binding affinities from ChEMBL to address selectivity among homologous carbonic anhydr...
-
Agentic AIs Are the Missing Paradigm for Out-of-Distribution Generalization in Foundation Models
Agentic AI systems are required to overcome the parameter coverage ceiling that prevents foundation models from handling certain out-of-distribution cases.
-
ReconVLA: An Uncertainty-Guided and Failure-Aware Vision-Language-Action Framework for Robotic Control
ReconVLA enhances pretrained vision-language-action robotic policies with conformal prediction for uncertainty estimation and failure detection without retraining.
-
OOM-RL: Out-of-Money Reinforcement Learning Market-Driven Alignment for LLM-Based Multi-Agent Systems
OOM-RL aligns multi-agent LLM systems for software engineering by using real financial market losses as an un-hackable negative gradient, resulting in a mature-phase annualized Sharpe ratio of 2.06 via a strict test-d...
-
Out-of-Distribution Generalization in Time Series: A Survey
This is the first comprehensive survey of OOD generalization methodologies for time series, organized across data distribution, representation learning, and OOD evaluation.
-
A Model-Driven Approach for Developing Families of Reinforcement Learning Environments
A hybrid genetic algorithm with model transformations generates families of RL training environments, demonstrated for wildfire mitigation and curriculum learning.
-
Towards Auditing AI Systems in the Wild
Proposes framing auditing of deployed AI systems as continuous statistical monitoring of risk-controlled constraints like fairness and safety under uncertainty.
-
Outsmarting the Chameleon: Counterfactual Decoupling for Tactical OOD Shifts in Live Streaming Risk Assessment
LPCD is a latent-predictive counterfactual decoupling framework that models intent and narrative variation separately at the latent level to improve risk prediction under adversarial tactical shifts in live streaming.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.