7 papers · 1 filter
The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations
Tyler LaBonte, John C. Hill, Xinchen Zhang +2
Modern machine learning models are prone to over-reliance on spurious correlations, which can often lead to poor performance on minority groups. In this paper, we identify surprisi…
Just Wing It: Near-Optimal Estimation of Missing Mass in a Markovian Sequence
Ashwin Pananjady, Vidya Muthukumar, Andrew Thangaraj
We study the problem of estimating the stationary mass -- also called the unigram mass -- that is missing from a single trajectory of a discrete-time, ergodic Markov chain. This pr…
Precise asymptotics of reweighted least-squares algorithms for linear diagonal networks
Chiraag Kaushik, Justin Romberg, Vidya Muthukumar
The classical iteratively reweighted least-squares (IRLS) algorithm aims to recover an unknown signal from linear measurements by performing a sequence of weighted least squares pr…
Balanced Data, Imbalanced Spectra: Unveiling Class Disparities with Spectral Imbalance
Chiraag Kaushik, Ran Liu, Chi-Heng Lin +5
Classification models are expected to perform equally well for different classes, yet in practice, there are often large gaps in their performance. This issue of class bias is wide…
Sharp analysis of out-of-distribution error for "importance-weighted" estimators in the overparameterized regime
Kuo-Wei Lai, Vidya Muthukumar
Overparameterized models that achieve zero training error are observed to generalize well on average, but degrade in performance when faced with data that is under-represented in t…
Faster Margin Maximization Rates for Generic and Adversarially Robust Optimization Methods
Guanghui Wang, Zihao Hu, Claudio Gentile +2
First-order optimization methods tend to inherently favor certain solutions over others when minimizing an underdetermined training objective that has multiple global optima. This…