Preetum Nakkiran
Harvard University
H-index: 18
North America-United States
Top articles of Preetum Nakkiran
Title | Journal | Author(s) | Publication Date |
---|---|---|---|
When Does Optimizing a Proper Loss Yield Calibration? | arXiv preprint arXiv:2305.18764 | Jarosław Błasiok Parikshit Gopalan Lunjia Hu Preetum Nakkiran | 2023/5/30 |
Loss Minimization Yields Multicalibration for Large Neural Networks | arXiv preprint arXiv:2304.09424 | Jarosław Błasiok Parikshit Gopalan Lunjia Hu Adam Tauman Kalai Preetum Nakkiran | 2023/4/19 |
What algorithms can transformers learn? a study in length generalization | arXiv preprint arXiv:2310.16028 | Hattie Zhou Arwen Bradley Etai Littwin Noam Razin Omid Saremi | 2023/10/24 |
Smooth ECE: Principled reliability diagrams via kernel smoothing | arXiv preprint arXiv:2309.12236 | Jarosław Błasiok Preetum Nakkiran | 2023/9/21 |
A unifying theory of distance from calibration | Jarosław Błasiok Parikshit Gopalan Lunjia Hu Preetum Nakkiran | 2023/6/2 | |
Perspectives on the State and Future of Deep Learning--2023 | arXiv preprint arXiv:2312.09323 | Micah Goldblum Anima Anandkumar Richard Baraniuk Tom Goldstein Kyunghyun Cho | 2023/12/7 |
Empirical limitations of the NTK for understanding scaling laws in deep learning | Transactions on Machine Learning Research | Nikhil Vyas Yamini Bansal Preetum Nakkiran | 2023/3/27 |
LiDAR: Sensing Linear Probing Performance in Joint Embedding SSL Architectures | arXiv preprint arXiv:2312.04000 | Vimal Thilak Chen Huang Omid Saremi Laurent Dinh Hanlin Goh | 2023/12/7 |
Vanishing gradients in reinforcement finetuning of language models | arXiv preprint arXiv:2310.20703 | Noam Razin Hattie Zhou Omid Saremi Vimal Thilak Arwen Bradley | 2023/10/31 |
Benign, tempered, or catastrophic: Toward a refined taxonomy of overfitting | Advances in Neural Information Processing Systems | Neil Mallinar James Simon Amirhesam Abedsoltan Parthe Pandit Misha Belkin | 2022/12/6 |
Limitations of neural collapse for understanding generalization in deep learning | arXiv preprint arXiv:2202.08384 | Like Hui Mikhail Belkin Preetum Nakkiran | 2022/2/17 |
APE: Aligning Pretrained Encoders to Quickly Learn Aligned Multimodal Representations | NeurIPS Workshop Has it Trained Yet? | Elan Rosenfeld Preetum Nakkiran Hadi Pouransari Oncel Tuzel Fartash Faghri | 2022/10/8 |
Near-Optimal NP-Hardness of Approximating Max -CSP | Theory of Computing | Pasin Manurangsi Preetum Nakkiran Luca Trevisan | 2022/2/14 |
The calibration generalization gap | arXiv preprint arXiv:2210.01964 | A Michael Carrell Neil Mallinar James Lucas Preetum Nakkiran | 2022/10/5 |
Knowledge Distillation: Bad Models Can Be Good Role Models | Advances in Neural Information Processing Systems | Gal Kaplun Eran Malach Preetum Nakkiran Shai Shalev-Shwartz | 2022/12/6 |
Limitations of the ntk for understanding generalization in deep learning | arXiv preprint arXiv:2206.10012 | Nikhil Vyas Yamini Bansal Preetum Nakkiran | 2022/6/20 |
Incentivizing empirical science in machine learning: Problems and proposals | ML Evaluation Standards Workshop at ICLR | Preetum Nakkiran Mikhail Belkin | 2022 |
General strong polarization | ACM Journal of the ACM (JACM) | Jarosław Błasiok Venkatesan Guruswami Preetum Nakkiran Atri Rudra Madhu Sudan | 2022/3/3 |
What you see is what you get: Principled deep learning via distributional generalization | Advances in Neural Information Processing Systems | Bogdan Kulynych Yao-Yuan Yang Yaodong Yu Jarosław Błasiok Preetum Nakkiran | 2022/12/6 |
Deconstructing distributions: A pointwise framework of learning | Gal Kaplun Nikhil Ghosh Saurabh Garg Boaz Barak Preetum Nakkiran | 2022/2/20 |