Nikunj Saunshi

Publication Activity (10 Years)

Years Active: 2017-2024
Publications (10 Years): 31

Top Topics

Sample Complexity

Top Venues

Publications

Abhishek Panigrahi, Nikunj Saunshi, Kaifeng Lyu, Sobhan Miryoosefi, Sashank J. Reddi, Satyen Kale, Sanjiv Kumar
Efficient Stagewise Pretraining via Progressive Subnetworks. CoRR (2024)
Stefani Karp, Nikunj Saunshi, Sobhan Miryoosefi, Sashank J. Reddi, Sanjiv Kumar
Landscape-Aware Growing: The Power of a Little LAG. CoRR (2024)
Vedant Gaur, Nikunj Saunshi
Reasoning in Large Language Models Through Symbolic Math Word Problems. ACL (Findings) (2023)
Vedant Gaur, Nikunj Saunshi
Reasoning in Large Language Models Through Symbolic Math Word Problems. CoRR (2023)
Nikunj Saunshi, Arushi Gupta, Mark Braverman, Sanjeev Arora
Understanding Influence Functions and Datamodels via Harmonic Analysis. ICLR (2023)
Abhishek Panigrahi, Nikunj Saunshi, Haoyu Zhao, Sanjeev Arora
Task-Specific Skill Localization in Fine-tuned Language Models. CoRR (2023)
Abhishek Panigrahi, Nikunj Saunshi, Haoyu Zhao, Sanjeev Arora
Task-Specific Skill Localization in Fine-tuned Language Models. ICML (2023)
Nikunj Saunshi, Jordan T. Ash, Surbhi Goel, Dipendra Misra, Cyril Zhang, Sanjeev Arora, Sham M. Kakade, Akshay Krishnamurthy
Understanding Contrastive Learning Requires Incorporating Inductive Biases. CoRR (2022)
Yi Zhang, Arushi Gupta, Nikunj Saunshi, Sanjeev Arora
On Predicting Generalization using GANs. ICLR (2022)
Arushi Gupta, Nikunj Saunshi, Dingli Yu, Kaifeng Lyu, Sanjeev Arora
New Definitions and Evaluations for Saliency Methods: Staying Intrinsic, Complete and Sound. CoRR (2022)
Arushi Gupta, Nikunj Saunshi, Dingli Yu, Kaifeng Lyu, Sanjeev Arora
New Definitions and Evaluations for Saliency Methods: Staying Intrinsic, Complete and Sound. NeurIPS (2022)
Nikunj Saunshi, Jordan T. Ash, Surbhi Goel, Dipendra Misra, Cyril Zhang, Sanjeev Arora, Sham M. Kakade, Akshay Krishnamurthy
Understanding Contrastive Learning Requires Incorporating Inductive Biases. ICML (2022)
Nikunj Saunshi, Arushi Gupta, Mark Braverman, Sanjeev Arora
Understanding Influence Functions and Datamodels via Harmonic Analysis. CoRR (2022)
Nikunj Saunshi, Arushi Gupta, Wei Hu
A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learning. ICML (2021)
Nikunj Saunshi, Arushi Gupta, Wei Hu
A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learning. CoRR (2021)
Nikunj Saunshi, Sadhika Malladi, Sanjeev Arora
A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks. ICLR (2021)
Jason D. Lee, Qi Lei, Nikunj Saunshi, Jiacheng Zhuo
Predicting What You Already Know Helps: Provable Self-Supervised Learning. NeurIPS (2021)
Yi Zhang, Arushi Gupta, Nikunj Saunshi, Sanjeev Arora
On Predicting Generalization using GANs. CoRR (2021)
Sanjeev Arora, Simon S. Du, Sham M. Kakade, Yuping Luo, Nikunj Saunshi
Provable Representation Learning for Imitation Learning via Bi-level Optimization. ICML (2020)
Jason D. Lee, Qi Lei, Nikunj Saunshi, Jiacheng Zhuo
Predicting What You Already Know Helps: Provable Self-Supervised Learning. CoRR (2020)
Nikunj Saunshi, Sadhika Malladi, Sanjeev Arora
A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks. CoRR (2020)
Nikunj Saunshi, Yi Zhang, Mikhail Khodak, Sanjeev Arora
A Sample Complexity Separation between Non-Convex and Convex Meta-Learning. ICML (2020)
Sanjeev Arora, Simon S. Du, Sham M. Kakade, Yuping Luo, Nikunj Saunshi
Provable Representation Learning for Imitation Learning via Bi-level Optimization. CoRR (2020)
Nikunj Saunshi, Yi Zhang, Mikhail Khodak, Sanjeev Arora
A Sample Complexity Separation between Non-Convex and Convex Meta-Learning. CoRR (2020)
Sanjeev Arora, Hrishikesh Khandeparkar, Mikhail Khodak, Orestis Plevrakis, Nikunj Saunshi
A Theoretical Analysis of Contrastive Unsupervised Representation Learning. CoRR (2019)
Nikunj Saunshi, Orestis Plevrakis, Sanjeev Arora, Mikhail Khodak, Hrishikesh Khandeparkar
A Theoretical Analysis of Contrastive Unsupervised Representation Learning. ICML (2019)
Mikhail Khodak, Nikunj Saunshi, Yingyu Liang, Tengyu Ma, Brandon Stewart, Sanjeev Arora
A La Carte Embedding: Cheap but Effective Induction of Semantic Feature Vectors. ACL (1) (2018)
Sanjeev Arora, Mikhail Khodak, Nikunj Saunshi, Kiran Vodrahalli
A Compressed Sensing View of Unsupervised Text Embeddings, Bag-of-n-Grams, and LSTMs. ICLR (Poster) (2018)
Mikhail Khodak, Nikunj Saunshi, Kiran Vodrahalli
A Large Self-Annotated Corpus for Sarcasm. LREC (2018)
Mikhail Khodak, Nikunj Saunshi, Yingyu Liang, Tengyu Ma, Brandon Stewart, Sanjeev Arora
A La Carte Embedding: Cheap but Effective Induction of Semantic Feature Vectors. CoRR (2018)
Mikhail Khodak, Nikunj Saunshi, Kiran Vodrahalli
A Large Self-Annotated Corpus for Sarcasm. CoRR (2017)