upper bound
An upper bound in AI refers to a theoretical limit on the performance or efficiency of an algorithm or model. It provides insights into the potential maximum capabilities of a system as it approaches certain constraints or optimality conditions.
- Agnostic Learning under Targeted Poisoning: Optimal Rates and the Role of Randomness
- Convergence of the Gradient Flow for Shallow ReLU Networks on Weakly Interacting Data
- Depth-Bounds for Neural Networks via the Braid Arrangement
- Depth-Bounds for Neural Networks via the Braid Arrangement
- Minimum Width for Deep, Narrow MLP: A Diffeomorphism Approach
- Optimal Rates in Continual Linear Regression via Increasing Regularization
- Price of Parsimony: Complexity of Fourier Sparsity Testing
- Prompt Tuning Transformers for Data Memorization
- RidgeLoRA: Matrix Ridge Enhanced Low-Rank Adaptation of Large Language Models
- Tight Bounds on the Distortion of Randomized and Deterministic Distributed Voting
- Tight Lower Bounds and Improved Convergence in Performative Prediction
- Variational Uncertainty Decomposition for In-Context Learning