Preetum Nakkiran

Publication Activity (10 Years)

Years Active: 2014-2024
Publications (10 Years): 60

Top Topics

Top Venues

Publications

Preetum Nakkiran, Arwen Bradley, Hattie Zhou, Madhu Advani
Step-by-Step Diffusion: An Elementary Tutorial. CoRR (2024)
Jaroslaw Blasiok, Parikshit Gopalan, Lunjia Hu, Adam Tauman Kalai, Preetum Nakkiran
Loss Minimization Yields Multicalibration for Large Neural Networks. ITCS (2024)
Noam Razin, Hattie Zhou, Omid Saremi, Vimal Thilak, Arwen Bradley, Preetum Nakkiran, Joshua M. Susskind, Etai Littwin
Vanishing Gradients in Reinforcement Finetuning of Language Models. ICLR (2024)
Vimal Thilak, Chen Huang, Omid Saremi, Laurent Dinh, Hanlin Goh, Preetum Nakkiran, Joshua M. Susskind, Etai Littwin
LiDAR: Sensing Linear Probing Performance in Joint Embedding SSL Architectures. ICLR (2024)
Hattie Zhou, Arwen Bradley, Etai Littwin, Noam Razin, Omid Saremi, Joshua M. Susskind, Samy Bengio, Preetum Nakkiran
What Algorithms can Transformers Learn? A Study in Length Generalization. ICLR (2024)
Dutch Hansen, Siddartha Devic, Preetum Nakkiran, Vatsal Sharan
When is Multicalibration Post-Processing Necessary? CoRR (2024)
Jaroslaw Blasiok, Preetum Nakkiran
Smooth ECE: Principled Reliability Diagrams via Kernel Smoothing. ICLR (2024)
Etai Littwin, Omid Saremi, Madhu Advani, Vimal Thilak, Preetum Nakkiran, Chen Huang, Joshua M. Susskind
How JEPA Avoids Noisy Features: The Implicit Bias of Deep Linear Self Distillation Networks. CoRR (2024)
Noam Razin, Hattie Zhou, Omid Saremi, Vimal Thilak, Arwen Bradley, Preetum Nakkiran, Joshua M. Susskind, Etai Littwin
Vanishing Gradients in Reinforcement Finetuning of Language Models. CoRR (2023)
Vimal Thilak, Chen Huang, Omid Saremi, Laurent Dinh, Hanlin Goh, Preetum Nakkiran, Joshua M. Susskind, Etai Littwin
LiDAR: Sensing Linear Probing Performance in Joint Embedding SSL Architectures. CoRR (2023)
Jaroslaw Blasiok, Parikshit Gopalan, Lunjia Hu, Adam Tauman Kalai, Preetum Nakkiran
Loss minimization yields multicalibration for large neural networks. CoRR (2023)
Nikhil Vyas, Yamini Bansal, Preetum Nakkiran
Empirical Limitations of the NTK for Understanding Scaling Laws in Deep Learning. Trans. Mach. Learn. Res. 2023 (2023)
Jaroslaw Blasiok, Parikshit Gopalan, Lunjia Hu, Preetum Nakkiran
A Unifying Theory of Distance from Calibration. STOC (2023)
Jaroslaw Blasiok, Parikshit Gopalan, Lunjia Hu, Preetum Nakkiran
When Does Optimizing a Proper Loss Yield Calibration? NeurIPS (2023)
Micah Goldblum, Anima Anandkumar, Richard G. Baraniuk, Tom Goldstein, Kyunghyun Cho, Zachary C. Lipton, Melanie Mitchell, Preetum Nakkiran, Max Welling, Andrew Gordon Wilson
Perspectives on the State and Future of Deep Learning - 2023. CoRR (2023)
Hattie Zhou, Arwen Bradley, Etai Littwin, Noam Razin, Omid Saremi, Josh M. Susskind, Samy Bengio, Preetum Nakkiran
What Algorithms can Transformers Learn? A Study in Length Generalization. CoRR (2023)
Jaroslaw Blasiok, Preetum Nakkiran
Smooth ECE: Principled Reliability Diagrams via Kernel Smoothing. CoRR (2023)
Gal Kaplun, Nikhil Ghosh, Saurabh Garg, Boaz Barak, Preetum Nakkiran
Deconstructing Distributions: A Pointwise Framework of Learning. ICLR (2023)
Jaroslaw Blasiok, Parikshit Gopalan, Lunjia Hu, Preetum Nakkiran
When Does Optimizing a Proper Loss Yield Calibration? CoRR (2023)
Gal Kaplun, Eran Malach, Preetum Nakkiran, Shai Shalev-Shwartz
Knowledge Distillation: Bad Models Can Be Good Role Models. CoRR (2022)
Neil Mallinar, James B. Simon, Amirhesam Abedsoltan, Parthe Pandit, Mikhail Belkin, Preetum Nakkiran
Benign, Tempered, or Catastrophic: A Taxonomy of Overfitting. CoRR (2022)
Bogdan Kulynych, Yao-Yuan Yang, Yaodong Yu, Jaroslaw Blasiok, Preetum Nakkiran
What You See is What You Get: Distributional Generalization for Algorithm Design in Deep Learning. CoRR (2022)
A. Michael Carrell, Neil Mallinar, James Lucas, Preetum Nakkiran
The Calibration Generalization Gap. CoRR (2022)
Jaroslaw Blasiok, Parikshit Gopalan, Lunjia Hu, Preetum Nakkiran
A Unifying Theory of Distance from Calibration. CoRR (2022)
Jaroslaw Blasiok, Venkatesan Guruswami, Preetum Nakkiran, Atri Rudra, Madhu Sudan
General Strong Polarization. J. ACM 69 (2) (2022)
Pasin Manurangsi, Preetum Nakkiran, Luca Trevisan
. Theory Comput. 18 (2022)
Like Hui, Mikhail Belkin, Preetum Nakkiran
Limitations of Neural Collapse for Understanding Generalization in Deep Learning. CoRR (2022)
Gal Kaplun, Nikhil Ghosh, Saurabh Garg, Boaz Barak, Preetum Nakkiran
Deconstructing Distributions: A Pointwise Framework of Learning. CoRR (2022)
Neil Mallinar, James B. Simon, Amirhesam Abedsoltan, Parthe Pandit, Misha Belkin, Preetum Nakkiran
Benign, Tempered, or Catastrophic: Toward a Refined Taxonomy of Overfitting. NeurIPS (2022)
Nikhil Vyas, Yamini Bansal, Preetum Nakkiran
Limitations of the NTK for Understanding Generalization in Deep Learning. CoRR (2022)
Bogdan Kulynych, Yao-Yuan Yang, Yaodong Yu, Jaroslaw Blasiok, Preetum Nakkiran
What You See is What You Get: Principled Deep Learning via Distributional Generalization. NeurIPS (2022)
Gal Kaplun, Eran Malach, Preetum Nakkiran, Shai Shalev-Shwartz
Knowledge Distillation: Bad Models Can Be Good Role Models. NeurIPS (2022)
Elan Rosenfeld, Preetum Nakkiran, Hadi Pouransari, Oncel Tuzel, Fartash Faghri
APE: Aligning Pretrained Encoders to Quickly Learn Aligned Multimodal Representations. CoRR (2022)
Preetum Nakkiran, Behnam Neyshabur, Hanie Sedghi
The Deep Bootstrap Framework: Good Online Learners are Good Offline Generalizers. ICLR (2021)
Yamini Bansal, Preetum Nakkiran, Boaz Barak
Revisiting Model Stitching to Compare Neural Representations. CoRR (2021)
Yamini Bansal, Preetum Nakkiran, Boaz Barak
Revisiting Model Stitching to Compare Neural Representations. NeurIPS (2021)
Preetum Nakkiran
Turing-Universal Learners with Optimal Scaling Laws. CoRR (2021)
Preetum Nakkiran, Prayaag Venkat, Sham M. Kakade, Tengyu Ma
Optimal Regularization can Mitigate Double Descent. ICLR (2021)
Preetum Nakkiran
Learning Rate Annealing Can Provably Help Generalization, Even for Convex Problems. CoRR (2020)
Preetum Nakkiran, Gal Kaplun, Yamini Bansal, Tristan Yang, Boaz Barak, Ilya Sutskever
Deep Double Descent: Where Bigger Models and More Data Hurt. ICLR (2020)
Preetum Nakkiran, Prayaag Venkat, Sham M. Kakade, Tengyu Ma
Optimal Regularization Can Mitigate Double Descent. CoRR (2020)
Preetum Nakkiran, Yamini Bansal
Distributional Generalization: A New Kind of Generalization. CoRR (2020)
Preetum Nakkiran, Behnam Neyshabur, Hanie Sedghi
The Deep Bootstrap: Good Online Learners are Good Offline Generalizers. CoRR (2020)
Preetum Nakkiran, Gal Kaplun, Dimitris Kalimeris, Tristan Yang, Benjamin L. Edelman, Fred Zhang, Boaz Barak
SGD on Neural Networks Learns Functions of Increasing Complexity. CoRR (2019)
Preetum Nakkiran, Gal Kaplun, Yamini Bansal, Tristan Yang, Boaz Barak, Ilya Sutskever
Deep Double Descent: Where Bigger Models and More Data Hurt. CoRR (2019)
Preetum Nakkiran
More Data Can Hurt for Linear Regression: Sample-wise Double Descent. CoRR (2019)
Akshay Degwekar, Preetum Nakkiran, Vinod Vaikuntanathan
Computational Limitations in Robust Classification and Win-Win Results. COLT (2019)
Venkatesan Guruswami, Preetum Nakkiran, Madhu Sudan
Algorithmic Polarization for Hidden Markov Models. ITCS (2019)
Chi-Ning Chou, Zhixian Lei, Preetum Nakkiran
Norm with Constant Update Time. APPROX-RANDOM (2019)
Preetum Nakkiran
Adversarial Robustness May Be at Odds With Simplicity. CoRR (2019)
Dimitris Kalimeris, Gal Kaplun, Preetum Nakkiran, Benjamin L. Edelman, Tristan Yang, Boaz Barak, Haofeng Zhang
SGD on Neural Networks Learns Functions of Increasing Complexity. NeurIPS (2019)
Chi-Ning Chou, Zhixian Lei, Preetum Nakkiran
Norm with Constant Update Time. CoRR (2018)
Preetum Nakkiran, Jaroslaw Blasiok
The Generic Holdout: Preventing False-Discoveries in Adaptive Data Science. CoRR (2018)
Jaroslaw Blasiok, Venkatesan Guruswami, Preetum Nakkiran, Atri Rudra, Madhu Sudan
General Strong Polarization. Electron. Colloquium Comput. Complex. 25 (2018)
Jaroslaw Blasiok, Venkatesan Guruswami, Preetum Nakkiran, Atri Rudra, Madhu Sudan
General Strong Polarization. CoRR (2018)
Jaroslaw Blasiok, Venkatesan Guruswami, Preetum Nakkiran, Atri Rudra, Madhu Sudan
General strong polarization. STOC (2018)
Venkatesan Guruswami, Preetum Nakkiran, Madhu Sudan
Algorithmic Polarization for Hidden Markov Models. CoRR (2018)
Charalampos E. Tsourakakis, Michael Mitzenmacher, Jaroslaw Blasiok, Ben Lawson, Preetum Nakkiran, Vasileios Nakos
Predicting Positive and Negative Links with Noisy Queries: Theory & Practice. CoRR (2017)
Preetum Nakkiran, K. V. Rashmi, Kannan Ramchandran
Optimal systematic distributed storage codes with fast encoding. ISIT (2016)
Pasin Manurangsi, Preetum Nakkiran, Luca Trevisan
Near-Optimal UGC-hardness of Approximating Max k-CSP_R. APPROX-RANDOM (2016)
K. V. Rashmi, Preetum Nakkiran, Jingyan Wang, Nihar B. Shah, Kannan Ramchandran
Having Your Cake and Eating It Too: Jointly Optimal Erasure Codes for I/O, Storage, and Network-bandwidth. FAST (2015)
Preetum Nakkiran, Raziel Alvarez, Rohit Prabhavalkar, Carolina Parada
Compressing deep neural networks using a rank-constrained topology. INTERSPEECH (2015)
Pasin Manurangsi, Preetum Nakkiran, Luca Trevisan
Near-Optimal UGC-hardness of Approximating Max k-CSP_R. CoRR (2015)
Preetum Nakkiran, K. V. Rashmi, Kannan Ramchandran
Optimal Systematic Distributed Storage Codes with Fast Encoding. CoRR (2015)
Rohit Prabhavalkar, Raziel Alvarez, Carolina Parada, Preetum Nakkiran, Tara N. Sainath
Automatic gain control and multi-style training for robust small-footprint keyword spotting with deep neural networks. ICASSP (2015)
Preetum Nakkiran, Nihar B. Shah, K. V. Rashmi
Fundamental limits on communication for oblivious updates in storage networks. GLOBECOM (2014)
Preetum Nakkiran, Nihar B. Shah, K. V. Rashmi
Fundamental Limits on Communication for Oblivious Updates in Storage Networks. CoRR (2014)