← Search

Shreyas Malakarjun Patil

2 accepted papers

2023

Neural Sculpting: Uncovering hierarchically modular task structure in neural networks through pruning and network analysis

NeurIPS 2023poster

Natural target functions and tasks typically exhibit hierarchical modularity -- they can be broken down into simpler sub-functions that are organized in a hierarchy. Such sub-functions have two important features: they have a distinct set of inputs (input-separability) and they are reused as inputs…

Cited by 4SourcePDFScholar
2021

PHEW : Constructing Sparse Networks that Learn Fast and Generalize Well without Training Data

ICML 2021spotlight

Methods that sparsify a network at initialization are important in practice because they greatly improve the efficiency of both learning and inference. Our work is based on a recently proposed decomposition of the Neural Tangent Kernel (NTK) that has decoupled the dynamics of the training process in…