3 papers
cs.LG2026
Do Flatter Minima Drive Better Generalization? An Algorithmic Separation in Grokking
Mohnish Harwani
Flat loss landscapes have long been linked to better generalization in neural networks. However, its role as a causal mechanism for generalization is less established. Grokking pro…
cs.LG2026
Sequential Pretraining Favors Large Models
Mohnish Harwani, Yujia Zheng
Large neural networks often acquire capabilities that small models fail to learn. Does this stem from large models learning more representative features, or from being more robust…
cond-mat.mtrl-sci2024
Accelerating active learning materials discovery with FAIR data and workflows: a case study for alloy melting temperatures
Mohnish Harwani, Juan C. Verduzco, Brian H. Lee +1
Active learning (AL) is a powerful sequential optimization approach that has shown great promise in the discovery of new materials. However, a major challenge remains the acquisiti…