Statistical Learning
Non-asymptotic Convergence of Training Transformers for Next-token Prediction
NTP is limited, with existing studies focusing mainly on asymptotic performance. This paper provides a fine-grained non-asymptotic analysis of the training dynamics of a one-layer transformer consisting of a self-attention module followed by a feed-forward layer.
Dissecting the Failure of Invariant Learning on Graphs
To address this, we propose Cross-environment Intra-class Alignment (CIA), which explicitly eliminates spurious features by aligning cross-environment representations conditioned on the same class, bypassing the need for explicit knowledge of the causal pattern structure.
Langevin Unlearning: A New Perspective of Noisy Gradient Descent for Machine Unlearning
Machine unlearning has raised significant interest with the adoption of laws ensuring the "right to be forgotten".
Semantic Feature Learning for Universal Unsupervised Cross-Domain Retrieval
However, existing efforts have several major limitations, with the most critical being their reliance on accurate supervision.
Explaining Datasets in Words: Statistical Models with Natural Language Parameters Ruiqi Zhong Heng Wang Dan Klein Jacob Steinhardt
However, these parameters are often high-dimensional and hard to interpret.
