Image de couverture de livre

Transformers Provably Learn Sparse XOR with Polylogarithmic Parameters

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Han, Yaomengxi, Ghoshdastidar, Debarghya
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

On the Convergence of Gradient Descent for Large Learning Rates
par: Crăciun, Alexandru, et autres
Publié: (2024)

Tight PAC-Bayesian Risk Certificates for Contrastive Learning
par: Van Elst, Anna, et autres
Publié: (2024)

Gaussian Process Limit Reveals Structural Benefits of Graph Transformers
par: Ayday, Nil, et autres
Publié: (2026)

Provable Robustness of (Graph) Neural Networks Against Data Poisoning and Backdoor Attacks
par: Gosch, Lukas, et autres
Publié: (2024)

Non-Singularity of the Gradient Descent map for Neural Networks with Piecewise Analytic Activations
par: Crăciun, Alexandru, et autres
Publié: (2025)

A Probabilistic Model for Non-Contrastive Learning
par: Fleissner, Maximilian, et autres
Publié: (2025)

Theoretical Foundations of Representation Learning using Unlabeled Data: Statistics and Optimization
par: Esser, Pascal, et autres
Publié: (2025)

A Theoretical Characterization of Optimal Data Augmentations in Self-Supervised Learning
par: Feigin, Shlomo Libo, et autres
Publié: (2024)

Impact of Bottleneck Layers and Skip Connections on the Generalization of Linear Denoising Autoencoders
par: Ham, Jonghyun, et autres
Publié: (2025)

Exact Generalisation Error Exposes Benchmarks Skew Graph Neural Networks Success (or Failure)
par: Ayday, Nil, et autres
Publié: (2025)

Recovering Imbalanced Clusters via Gradient-Based Projection Pursuit
par: Eppert, Martin, et autres
Publié: (2025)

Explainable Clustering Beyond Worst-Case Guarantees
par: Fleissner, Maximilian, et autres
Publié: (2024)

Interpretable Self-Supervised Learning via Representer Landmarks and Nyström Approximation
par: Zarvandi, Maedeh, et autres
Publié: (2025)

When can we Approximate Wide Contrastive Models with Neural Tangent Kernels and Principal Component Analysis?
par: Anil, Gautham Govind, et autres
Publié: (2024)

Explaining Kernel Clustering via Decision Trees
par: Fleissner, Maximilian, et autres
Publié: (2024)

Infinite Width Limits of Self Supervised Neural Networks
par: Fleissner, Maximilian, et autres
Publié: (2024)

Nonparametric Kernel Clustering with Bandit Feedback
par: Thuot, Victor, et autres
Publié: (2026)

Robustness Certificates for Neural Networks against Adversarial Attacks
par: Taheri, Sara, et autres
Publié: (2025)

Exact Certification of (Graph) Neural Networks Against Label Poisoning
par: Sabanayagam, Mahalakshmi, et autres
Publié: (2024)

Robust Feature Inference: A Test-time Defense Strategy using Spectral Projections
par: Singh, Anurag, et autres
Publié: (2023)

Generalization Certificates for Adversarially Robust Bayesian Linear Regression
par: Sabanayagam, Mahalakshmi, et autres
Publié: (2025)

Different Statistical Perspectives for Understanding Generalisation in Graph Neural Networks
par: Ayday, Nil, et autres
Publié: (2026)

Exact Certification of Neural Networks and Partition Aggregation Ensembles against Label Poisoning
par: Mohgaonkar, Ajinkya, et autres
Publié: (2026)

Learning on Transformers is Provable Low-Rank and Sparse: A One-layer Analysis
par: Li, Hongkang, et autres
Publié: (2024)

Transformers Provably Learn Sparse Token Selection While Fully-Connected Nets Cannot
par: Wang, Zixuan, et autres
Publié: (2024)

Provable In-Context Learning of Nonlinear Regression with Transformers
par: Li, Hongbo, et autres
Publié: (2025)

Transformers Provably Learn to Internalize Chain-of-Thought
par: Huang, Yixiao, et autres
Publié: (2026)

Provably Transformers Harness Multi-Concept Word Semantics for Efficient In-Context Learning
par: Bu, Dake, et autres
Publié: (2024)

Adaptive Estimation and Inference in Semi-parametric Heterogeneous Clustered Multitask Learning via Neyman Orthogonality
par: Chen, Hanxiao, et autres
Publié: (2026)

Unlabeled Data Can Provably Enhance In-Context Learning of Transformers
par: Liu, Renpu, et autres
Publié: (2026)

Looped Transformers with Layer Normalization Provably Learn the Power Method
par: Wu, Lyumin, et autres
Publié: (2026)

Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training
par: Bu, Dake, et autres
Publié: (2025)

Compute Optimal Inference and Provable Amortisation Gap in Sparse Autoencoders
par: O'Neill, Charles, et autres
Publié: (2024)

Sparse Orthogonal Parameters Tuning for Continual Learning
par: Ning, Kun-Peng, et autres
Publié: (2024)

Neuron Block Dynamics for XOR Classification with Zero-Margin
par: Braun, Guillaume, et autres
Publié: (2026)

Semi-Supervised Sparse Gaussian Classification: Provable Benefits of Unlabeled Data
par: Azar, Eyar, et autres
Publié: (2024)

Taming the Sigmoid Bottleneck: Provably Argmaxable Sparse Multi-Label Classification
par: Grivas, Andreas, et autres
Publié: (2023)

One-Layer Transformer Provably Learns One-Nearest Neighbor In Context
par: Li, Zihao, et autres
Publié: (2024)

Transformers Provably Learn Algorithmic Solutions for Graph Connectivity, But Only with the Right Data
par: Ye, Qilin, et autres
Publié: (2025)

Transformers Provably Implement In-Context Reinforcement Learning with Policy Improvement
par: Liang, Haodong, et autres
Publié: (2026)