| The merged-staircase property: a necessary and nearly sufficient condition for sgd learning of sparse functions on two-layer neural networks E Abbe, EB Adsera, T Misiakiewicz Conference on Learning Theory, 4782-4887, 2022 | 121 | 2022 |
| Sgd learning on neural networks: leap complexity and saddle-to-saddle dynamics E Abbe, EB Adsera, T Misiakiewicz The Thirty Sixth Annual Conference on Learning Theory, 2552-2623, 2023 | 80 | 2023 |
| The staircase property: How hierarchical structure can guide deep learning E Abbe, E Boix-Adsera, MS Brennan, G Bresler, D Nagaraj Advances in Neural Information Processing Systems 34, 26989-27002, 2021 | 64 | 2021 |
| Wasserstein barycenters are NP-hard to compute JM Altschuler, E Boix-Adsera SIAM Journal on Mathematics of Data Science 4 (1), 179-203, 2022 | 59 | 2022 |
| Wasserstein barycenters can be computed in polynomial time in fixed dimension JM Altschuler, E Boix-Adsera Journal of Machine Learning Research 22 (44), 1-19, 2021 | 53 | 2021 |
| Sample efficient active learning of causal trees K Greenewald, D Katz, K Shanmugam, S Magliacane, M Kocaoglu, ... Advances in Neural Information Processing Systems 32, 2019 | 48 | 2019 |
| The Average-Case Complexity of Counting Cliques in Erdös--Rényi Hypergraphs E Boix-Adserà, M Brennan, G Bresler SIAM Journal on Computing, FOCS19-39-FOCS19-80, 2021 | 47 | 2021 |
| Graph powering and spectral robustness E Abbe, E Boix-Adserà, P Ralli, C Sandon SIAM Journal on Mathematics of Data Science 2 (1), 132-157, 2020 | 43 | 2020 |
| Polynomial-time algorithms for multimarginal optimal transport problems with structure JM Altschuler, E Boix-Adsera Mathematical Programming 199 (1-2), 1107-1178, 2023 | 42 | 2023 |
| The multiplayer colonel blotto game E Boix-Adserà, BL Edelman, S Jayanti Games and Economic Behavior 129, 15-31, 2021 | 36 | 2021 |
| Hardness results for Multimarginal Optimal Transport problems JM Altschuler, E Boix-Adsera Discrete Optimization 42, 100669, 2021 | 35 | 2021 |
| Transformers learn through gradual rank increase E Boix-Adserà, E Littwin, E Abbe, S Bengio, JM Susskind Thirty-seventh Conference on Neural Information Processing Systems, 2023 | 27* | 2023 |
| Randomized concurrent set union and generalized wake-up S Jayanti, RE Tarjan, E Boix-Adserà Proceedings of the 2019 ACM Symposium on Principles of Distributed Computing …, 2019 | 21 | 2019 |
| On the non-universality of deep learning: quantifying the cost of symmetry E Abbe, E Boix-Adsera Advances in Neural Information Processing Systems 35, 17188-17201, 2022 | 19 | 2022 |
| Chow-liu++: Optimal prediction-centric learning of tree ising models E Boix-Adsera, G Bresler, F Koehler 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS …, 2022 | 14 | 2022 |
| GULP: a prediction-based metric between representations E Boix-Adsera, H Lawrence, G Stepaniants, P Rigollet Advances in Neural Information Processing Systems 35, 7115-7127, 2022 | 10 | 2022 |
| When can transformers reason with abstract symbols? E Boix-Adserà, O Saremi, E Abbe, S Bengio, E Littwin, J Susskind arXiv preprint arXiv:2310.09753, 2023 | 9 | 2023 |
| An information-percolation bound for spin synchronization on general graphs E Abbe, E Boix-Adserà Annals of Applied Probability 30 (3), 1066-1090, 2020 | 7 | 2020 |
| Towards a theory of model distillation E Boix-Adsera arXiv preprint arXiv:2403.09053, 2024 | 6 | 2024 |
| PROPANE: Prompt design as an inverse problem R Melamed, LH McCabe, T Wakhare, Y Kim, HH Huang, E Boix-Adsera arXiv preprint arXiv:2311.07064, 2023 | 4 | 2023 |
Jason M. AltschulerUPennVerified email at upenn.edu
