Takip et
Nikolaus Howe
Nikolaus Howe
PhD candidate at Mila, Université de Montréal
mila.quebec üzerinde doğrulanmış e-posta adresine sahip - Ana Sayfa
Başlık
Alıntı yapanlar
Alıntı yapanlar
Yıl
Defining and Characterizing Reward Hacking
J Skalse, N Howe, D Krasheninnikov, D Krueger
Advances in Neural Information Processing Systems 35, 9460-9471, 2022
3282022
Myriad: a real-world testbed to bridge trajectory optimization and deep learning
N Howe, S Dufort-Labbé, N Rajkumar, PL Bacon
Advances in Neural Information Processing Systems 35, 29801-29815, 2022
62022
Exploring Scaling Trends in LLM Robustness
N Howe, M Zając, I McKenzie, O Hollinsworth, PL Bacon, A Gleave
ICML 2024 Next Generation of AI Safety Workshop, 2024
52024
Scaling Trends in Language Model Robustness
N Howe, I McKenzie, O Hollinsworth, M Zajac, T Tseng, A Tucker, ...
arXiv preprint arXiv:2407.18213, 2025
3*2025
Learning neural ordinary differential equations for optimal control
NHR Howe
2022
Sistem, işlemi şu anda gerçekleştiremiyor. Daha sonra yeniden deneyin.
Makaleler 1–5
OSZAR »