Ansh Nagda
anshnagda (at) berkeley (dot) edu
I work on AI inference at Etched.
Previously I worked on AI+math at Google DeepMind (see below for some papers) and did my PhD in Theoretical Computer Science at UC Berkeley, focusing on complexity theory (also see below for some papers).
Papers
-
Reinforced Generation of Combinatorial Structures: Ramsey Numbers
Ansh Nagda, Prabhakar Raghavan, Abhradeep Thakurta
-
Reinforced Generation of Combinatorial Structures: Applications to Complexity Theory
Ansh Nagda, Prabhakar Raghavan, Abhradeep Thakurta
-
On optimal distinguishers for planted clique
Ansh Nagda, Prasad Raghavendra
FOCS 2025
-
Improved approximation algorithms for the EPR Hamiltonian
Nathan Ju, Ansh Nagda
APPROX 2025
-
On approximability of the Permanent of PSD matrices
Farzam Ebrahimnejad, Shayan Oveis Gharan, Ansh Nagda
STOC 2025
-
Counting and Sampling Perfect Matchings in Regular Expanding Non-Bipartite Graphs
Farzam Ebrahimnejad, Shayan Oveis Gharan, Ansh Nagda
ITCS 2022
-
Sublinear Time Hypergraph Sparsification via Cut and Edge Sampling Queries
Yu Chen, Sanjeev Khanna, Ansh Nagda
ICALP 2021
-
Near-linear Size Hypergraph Cut Sparsifiers
Yu Chen, Sanjeev Khanna, Ansh Nagda
FOCS 2020