| Invariance principle meets information bottleneck for out-of-distribution generalization K Ahuja, E Caballero, D Zhang, JC Gagnon-Audet, Y Bengio, I Mitliagkas, ... NeurIPS 2021 (Spotlight) 34, 3438-3450, 2021 | 403 | 2021 |
| Invariant risk minimization games K Ahuja, K Shanmugam, K Varshney, A Dhurandhar International Conference on Machine Learning, 145-155, 2020 | 354 | 2020 |
| Interventional causal representation learning K Ahuja, Y Wang, D Mahajan, Y Bengio ICML 2023 (Oral), arXiv preprint arXiv:2209.11924, 2022 | 180 | 2022 |
| Model ratatouille: Recycling diverse models for out-of-distribution generalization A Rame, K Ahuja, J Zhang, M Cord, L Bottou, D Lopez-Paz ICML 2023, 2023 | 145* | 2023 |
| Can subnetwork structure be the key to out-of-distribution generalization? D Zhang, K Ahuja, Y Xu, Y Wang, A Courville ICML 2021, Long Talk, 12356-12367, 2021 | 119 | 2021 |
| Empirical or invariant risk minimization? a sample complexity perspective K Ahuja, J Wang, A Dhurandhar, K Shanmugam, KR Varshney ICLR 2021, arXiv preprint arXiv:2010.16412, 2020 | 117 | 2020 |
| Weakly supervised representation learning with sparse perturbations K Ahuja, JS Hartford, Y Bengio Advances in Neural Information Processing Systems 35, 15516-15528, 2022 | 94 | 2022 |
| Sand-mask: An enhanced gradient masking strategy for the discovery of invariances in domain generalization S Shahtalebi, JC Gagnon-Audet, T Laleh, M Faramarzi, K Ahuja, I Rish arXiv preprint arXiv:2106.02266, 2021 | 76 | 2021 |
| Woods: Benchmarks for out-of-distribution generalization in time series JC Gagnon-Audet, K Ahuja, MJ Darvishi-Bayazi, P Mousavi, G Dumas, ... arXiv preprint arXiv:2203.09978, 2022 | 66 | 2022 |
| Properties from mechanisms: an equivariance perspective on identifiable representation learning K Ahuja, J Hartford, Y Bengio ICLR 2022 (Spotlight), arXiv preprint arXiv:2110.15796, 2021 | 44 | 2021 |
| A micro-foundation of social capital in evolving social networks AM Alaa, K Ahuja, M van der Schaar IEEE Transactions on Network Science and Engineering 5 (1), 14-31, 2017 | 31 | 2017 |
| Adversarial feature desensitization P Bashivan, R Bayat, A Ibrahim, K Ahuja, M Faramarzi, T Laleh, ... Advances in Neural Information Processing Systems 34, 10665-10677, 2021 | 30 | 2021 |
| Distributed interference management policies for heterogeneous small cell networks K Ahuja, Y Xiao, M van der Schaar IEEE Journal on Selected Areas in Communications 33 (6), 1112-1126, 2015 | 30 | 2015 |
| Multi-domain causal representation learning via weak distributional invariances K Ahuja, A Mansouri, Y Wang International Conference on Artificial Intelligence and Statistics, 865-873, 2024 | 24 | 2024 |
| Why does Throwing Away Data Improve Worst-Group Error? K Chaudhuri, K Ahuja, M Arjovsky, D Lopez-Paz ICML 2023, Oral, 2023 | 23 | 2023 |
| Dpscreen: Dynamic personalized screening K Ahuja, W Zame, M van der Schaar Advances in Neural Information Processing Systems 30, 2017 | 22 | 2017 |
| A closer look at in-context learning under distribution shifts K Ahuja, D Lopez-Paz arXiv preprint arXiv:2305.16704, 2023 | 21 | 2023 |
| Towards efficient representation identification in supervised learning K Ahuja, D Mahajan, V Syrgkanis, I Mitliagkas Conference on Causal Learning and Reasoning, 19-43, 2022 | 21 | 2022 |
| Fl games: A federated learning framework for distribution shifts S Gupta, K Ahuja, M Havaei, N Chatterjee, Y Bengio arXiv preprint arXiv:2205.11101, 2022 | 20 | 2022 |
| Estimating kullback-leibler divergence using kernel machines K Ahuja 2019 53rd Asilomar Conference on Signals, Systems, and Computers, 690-696, 2019 | 18 | 2019 |
Amit DhurandharPrincipal Research Scientist, IBMVerified email at us.ibm.com
