implicit bias
Implicit bias in AI refers to the unintended prejudices that can occur in algorithms due to biased training data or design choices. Addressing implicit bias is important for ensuring fairness and equity in AI applications.
- Convergence Rates for Gradient Descent on the Edge of Stability for Overparametrised Least Squares
- Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers
- Heavy-Ball Momentum Method in Continuous Time and Discretization Error Analysis
- Memorization in Graph Neural Networks
- Neural Collapse under Gradient Flow on Shallow ReLU Networks for Orthogonally Separable Data
- Stable Minima of ReLU Neural Networks Suffer from the Curse of Dimensionality: The Neural Shattering Phenomenon
- The Implicit Bias of Structured State Space Models Can Be Poisoned With Clean Labels
- The Rich and the Simple: On the Implicit Bias of Adam and SGD