Sotabase

Career

· AI Researcher, Google DeepMind2023–
· Postdoctoral Researcher, Stanford University2023–
· Education, University of California, Berkeley2016–2022
· PhD Student, University of California, Berkeley2016–2022
· Researcher, Microsoft Research India2014–2016
· Bachelor of Engineering, Indian Institute of Technology Delhi2010–2014

Publications (36)

Sparse Local Embeddings for Extreme Multi-label Classification
Neural Information Processing Systems · 2015
453
cited
The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
International Conference on Learning Representations · 2022
255
cited
Derivative-Free Methods for Policy Optimization: Guarantees for Linear Quadratic Systems
International Conference on Artificial Intelligence and Statistics · 2018
220
cited
FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network
Neural Information Processing Systems · 2018
198
cited
Robust Regression via Hard Thresholding
Neural Information Processing Systems · 2015
163
cited
POLITEX: Regret Bounds for Policy Iteration using Expert Prediction
International Conference on Machine Learning · 2019
136
cited
Establishing Appropriate Trust via Critical States
IEEE/RJS International Conference on Intelligent RObots and Systems · 2018
125
cited
Consistent Robust Regression
Neural Information Processing Systems · 2017
104
cited
Skill-it! A Data-Driven Skills Framework for Understanding and Training Language Models
Neural Information Processing Systems · 2023
96
cited
Adaptive Hard Thresholding for Near-optimal Consistent Robust Regression
Annual Conference Computational Learning Theory · 2019
51
cited
On the Sensitivity of Reward Inference to Misspecified Human Models
International Conference on Learning Representations · 2022
28
cited
Gen-Oja: Simple & Efficient Algorithm for Streaming Generalized Eigenvector Computation
Neural Information Processing Systems · 2018
17
cited
TART: A plug-and-play Transformer module for task-agnostic reasoning
Neural Information Processing Systems · 2023
16
cited
Online learning with dynamics: A minimax perspective
Neural Information Processing Systems · 2020
15
cited
Preference learning along multiple criteria: A game-theoretic perspective
Neural Information Processing Systems · 2021
14
cited
Bayesian Robustness: A Nonasymptotic Viewpoint
Journal of the American Statistical Association · 2019
10
cited
A case for reframing automated medical image classification as segmentation
Neural Information Processing Systems · 2023
9
cited
Does In-Context Operator Learning Generalize to Domain-Shifted Settings?
8
cited
Lazy Generic Cuts
Computer Vision and Image Understanding · 2016
8
cited
Locally Non-linear Embeddings for Extreme Multi-label Learning
arXiv.org · 2015
8
cited
Sotabase
Kush Bhatia | Researcher Profile | Sotabase | Sotabase