Követés
Nikunj Saunshi
Nikunj Saunshi
Research Scientist, Google
E-mail megerősítve itt: google.com - Kezdőlap
Cím
Hivatkozott rá
Hivatkozott rá
Év
A Theoretical Analysis of Contrastive Unsupervised Representation Learning
S Arora, H Khandeparkar, M Khodak, O Plevrakis, N Saunshi
International Conference on Machine Learning (ICML) 2019, 2019
889*2019
A Large Self-Annotated Corpus for Sarcasm
M Khodak, N Saunshi, K Vodrahalli
Language Resources and Evaluation Conference (LREC) 2018, 2017
2902017
Predicting What You Already Know Helps: Provable Self-Supervised Learning
JD Lee, Q Lei, N Saunshi, J Zhuo
Neural Information Processing Systems (NeurIPS) 2021, 2020
2072020
A La Carte Embedding: Cheap but Effective Induction of Semantic Feature Vectors
M Khodak*, N Saunshi*, Y Liang, T Ma, B Stewart, S Arora
Association for Computational Linguistics (ACL) 2018, 2018
1322018
Understanding Contrastive Learning Requires Incorporating Inductive Biases
N Saunshi, J Ash, S Goel, D Misra, C Zhang, S Arora, S Kakade, ...
International Conference on Machine Learning (ICML) 2022, 2022
1252022
A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks
N Saunshi, S Malladi, S Arora
International Conference on Learning Representations (ICLR) 2021, 2020
892020
Provable representation learning for imitation learning via bi-level optimization
S Arora, S Du, S Kakade, Y Luo, N Saunshi
International Conference on Machine Learning (ICML) 2020, 2020
762020
Task-Specific Skill Localization in Fine-tuned Language Models
A Panigrahi*, N Saunshi*, H Zhao, S Arora
International Conference on Machine Learning (ICML) 2023, 2023
692023
A compressed sensing view of unsupervised text embeddings, bag-of-n-grams, and LSTMs
S Arora, M Khodak, N Saunshi, K Vodrahalli
International Conference on Learning Representations (ICLR) 2018, 2018
542018
Reasoning in Large Language Models Through Symbolic Math Word Problems
V Gaur, N Saunshi
Findings of the Association for Computational Linguistics (ACL) 2023, 2023
31*2023
A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learning
N Saunshi, A Gupta, W Hu
International Conference on Machine Learning (ICML) 2021, 2021
262021
Understanding Influence Functions and Datamodels via Harmonic Analysis
N Saunshi, A Gupta, M Braverman, S Arora
International Conference on Learning Representations (ICLR) 2023, 2022
252022
A sample complexity separation between non-convex and convex meta-learning
N Saunshi, Y Zhang, M Khodak, S Arora
International Conference on Machine Learning (ICML) 2020, 2020
242020
On Predicting Generalization using GANs
Y Zhang, A Gupta, N Saunshi, S Arora
International Conference on Learning Representations (ICLR) 2022, 2021
142021
Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning?
K Gatmiry, N Saunshi, SJ Reddi, S Jegelka, S Kumar
Forty-first International Conference on Machine Learning, 2024
132024
Pixie: a social chatbot
O Adewale, A Beatson, D Buniatyan, J Ge, M Khodak, H Lee, N Prasad, ...
Alexa prize proceedings, 2017
132017
New Definitions and Evaluations for Saliency Methods: Staying Intrinsic, Complete and Sound
A Gupta*, N Saunshi*, D Yu*, K Lyu, S Arora
Neural Information Processing Systems (NeurIPS) 2022, 2021
72021
Efficient Stagewise Pretraining via Progressive Subnetworks
A Panigrahi*, N Saunshi*, K Lyu, S Miryoosefi, S Reddi, S Kale, S Kumar
International Conference on Learning Representations (ICLR) 2025, 2024
52024
A Little Help Goes a Long Way: Efficient LLM Training by Leveraging Small LMs
AS Rawat, V Sadhanala, A Rostamizadeh, A Chakrabarti, W Jitkrittum, ...
arXiv preprint arXiv:2410.18779, 2024
22024
Landscape-Aware Growing: The Power of a Little LAG
S Karp*, N Saunshi*, S Miryoosefi, SJ Reddi, S Kumar
arXiv preprint arXiv:2406.02469, 2024
22024
A rendszer jelenleg nem tudja elvégezni a műveletet. Próbálkozzon újra később.
Cikkek 1–20