Image de couverture de livre

A Theoretical Analysis of Compositional Generalization in Neural Networks: A Necessary and Sufficient Condition

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Li, Yuanpeng
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

What Makes Data Suitable for a Locally Connected Neural Network? A Necessary and Sufficient Condition Based on Quantum Entanglement
par: Alexander, Yotam, et autres
Publié: (2023)

Sufficient and Necessary Explanations (and What Lies in Between)
par: Bharti, Beepul, et autres
Publié: (2024)

On the Detection of Commutative Factors in Factor Graphs: Necessary and Sufficient Conditions
par: Luttermann, Malte, et autres
Publié: (2026)

Causally Sufficient and Necessary Feature Expansion for Class-Incremental Learning
par: Zhang, Zhen, et autres
Publié: (2026)

Data-Centric Machine Learning for Earth Observation: Necessary and Sufficient Features
par: Najjar, Hiba, et autres
Publié: (2024)

Graph Neural Network Surrogates for Contacting Deformable Bodies with Necessary and Sufficient Contact Detection
par: Dubey, Vijay K., et autres
Publié: (2025)

Invariant Learning via Probability of Sufficient and Necessary Causes
par: Yang, Mengyue, et autres
Publié: (2023)

Sufficient Conditions for Stability of Minimum-Norm Interpolating Deep ReLU Networks
par: Harzli, Ouns El, et autres
Publié: (2026)

Conditional Generative Models are Sufficient to Sample from Any Causal Effect Estimand
par: Rahman, Md Musfiqur, et autres
Publié: (2024)

Attribution Explanations for Deep Neural Networks: A Theoretical Perspective
par: Deng, Huiqi, et autres
Publié: (2025)

Are Expressive Encoders Necessary for Discrete Graph Generation?
par: Revolinsky, Jay, et autres
Publié: (2026)

Compositional Function Networks: A High-Performance Alternative to Deep Neural Networks with Built-in Interpretability
par: Li, Fang
Publié: (2025)

Modeling the Data-Generating Process is Necessary for Out-of-Distribution Generalization
par: Kaur, Jivat Neet, et autres
Publié: (2022)

When Does Learning Renormalize? Sufficient Conditions for Power Law Spectral Dynamics
par: Zhang, Yizhou
Publié: (2025)

Are Human-generated Demonstrations Necessary for In-context Learning?
par: Li, Rui, et autres
Publié: (2023)

A Theoretical Analysis of Discrete Flow Matching Generative Models
par: Su, Maojiang, et autres
Publié: (2025)

A Theoretical Analysis of Soft-Label vs Hard-Label Training in Neural Networks
par: Mandal, Saptarshi, et autres
Publié: (2024)

Relational Composition in Neural Networks: A Survey and Call to Action
par: Wattenberg, Martin, et autres
Publié: (2024)

Are Expressive Models Truly Necessary for Offline RL?
par: Wang, Guan, et autres
Publié: (2024)

Generalization in VAE and Diffusion Models: A Unified Information-Theoretic Analysis
par: Chen, Qi, et autres
Publié: (2025)

Ray-Tracing for Conditionally Activated Neural Networks
par: Gallicchio, Claudio, et autres
Publié: (2025)

Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for Reasoning
par: Yao, Xinhao, et autres
Publié: (2025)

What makes Models Compositional? A Theoretical View: With Supplement
par: Ram, Parikshit, et autres
Publié: (2024)

Physics-Guided Multimodal Transformers are the Necessary Foundation for the Next Generation of Meteorological Science
par: Han, Jing, et autres
Publié: (2025)

Generalization in Neural Networks: A Broad Survey
par: Rohlfs, Chris
Publié: (2022)

Necessary and Sufficient Conditions for Optimal Decision Trees using Dynamic Programming
par: van der Linden, Jacobus G. M., et autres
Publié: (2023)

Is Monotonic Sampling Necessary in Diffusion Models?
par: Khan, Muhammad Haris
Publié: (2026)

Scalable Evaluation and Neural Models for Compositional Generalization
par: Camposampiero, Giacomo, et autres
Publié: (2025)

Action-Sufficient Goal Representations
par: Hyeon, Jinu, et autres
Publié: (2026)

Generative Conditional Distributions by Neural (Entropic) Optimal Transport
par: Nguyen, Bao, et autres
Publié: (2024)

A Generalized Singular Value Theory for Neural Networks
par: Brown, Brian Charles, et autres
Publié: (2026)

A Theoretical Analysis of Test-Driven Code Generation
par: Menet, Nicolas, et autres
Publié: (2026)

Is Optimal Transport Necessary for Inverse Reinforcement Learning?
par: Dong, Zixuan, et autres
Publié: (2025)

Neural Predictive Control to Coordinate Discrete- and Continuous-Time Models for Time-Series Analysis with Control-Theoretical Improvements
par: Li, Haoran, et autres
Publié: (2025)

Sufficient Invariant Learning for Distribution Shift
par: Kim, Taero, et autres
Publié: (2022)

Seeking the Sufficiency and Necessity Causal Features in Multimodal Representation Learning
par: Chen, Boyu, et autres
Publié: (2024)

MEASURE: Multi-scale Minimal Sufficient Representation Learning for Domain Generalization in Sleep Staging
par: Jo, Sangmin, et autres
Publié: (2025)

Anytime-Valid Answer Sufficiency Certificates for LLM Generation via Sequential Information Lift
par: Akter, Sanjeda, et autres
Publié: (2025)

Bridging Smoothness and Approximation: Theoretical Insights into Over-Smoothing in Graph Neural Networks
par: Yang, Guangrui, et autres
Publié: (2024)

Graph-Conditioned Mixture of Graph Neural Network Experts for Traffic Forecasting
par: Ghaffari, Amirhossein, et autres
Publié: (2026)