Theory

Beyond TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles AAAI 2024 Bridging the Divide: Reconsidering Softmax and Linear Attention NeurIPS 2024 Can Transformers Learn n-gram Language Models? EMNLP 2024 Changing the Training Data Distribution to Reduce Simplicity Bias Improves In-distribution Generalization NeurIPS 2024 Collaborative Performance Prediction for Large Language Models EMNLP 2024 Combining Statistical Depth and Fermat Distance for Uncertainty Quantification NeurIPS 2024 Compact Proofs of Model Performance via Mechanistic Interpretability NeurIPS 2024 Convergence of Message-Passing Graph Neural Networks with Generic Aggregation on Large Random Graphs JMLR 2024 Convolutional Spectral Kernel Learning with Generalization Guarantees (Abstract Reprint) AAAI 2024 Curvature Clues: Decoding Deep Learning Privacy with Input Loss Curvature NeurIPS 2024 DASH: Warm-Starting Neural Network Training in Stationary Settings without Loss of Plasticity NeurIPS 2024 DeepBern-Nets: Taming the Complexity of Certifying Neural Networks Using Bernstein Polynomial Activations and Precise Bound Propagation AAAI 2024 Deep Learning for Computing Convergence Rates of Markov Chains NeurIPS 2024 Deep linear networks for regression are implicitly regularized towards flat minima NeurIPS 2024 Deep Neural Networks via Complex Network Theory: A Perspective IJCAI 2024 Depth Separation in Norm-Bounded Infinite-Width Neural Networks COLT 2024 Detection and Measurement of Syntactic Templates in Generated Text EMNLP 2024 Dissecting the Interplay of Attention Paths in a Statistical Mechanics Theory of Transformers NeurIPS 2024 Does Large Language Model Contain Task-Specific Neurons? EMNLP 2024 Double-Descent Curves in Neural Networks: A New Perspective Using Gaussian Processes AAAI 2024 ECLipsE: Efficient Compositional Lipschitz Constant Estimation for Deep Neural Networks NeurIPS 2024 Effect of Ambient-Intrinsic Dimension Gap on Adversarial Vulnerability AISTATS 2024 Efficiently Computing Susceptibility to Context in Language Models EMNLP 2024 Efficiently Learning One-Hidden-Layer ReLU Networks via SchurPolynomials COLT 2024 Enriching Disentanglement: From Logical Definitions to Quantitative Metrics NeurIPS 2024