Seguir
Melrose Roderick
Melrose Roderick
Postdoc - Mila
Dirección de correo verificada de mila.quebec - Página principal
Título
Citado por
Citado por
Año
Dex-net 1.0: A cloud-based network of 3d objects for robust grasp planning using a multi-armed bandit model with correlated rewards
J Mahler, FT Pokorny, B Hou, M Roderick, M Laskey, M Aubry, K Kohlhoff, ...
2016 IEEE international conference on robotics and automation (ICRA), 1957-1964, 2016
4062016
Implementing the deep q-network
M Roderick, J MacGlashan, S Tellex
arXiv preprint arXiv:1711.07478, 2017
682017
Enforcing robust control guarantees within neural network policies
PL Donti, M Roderick, M Fazlyab, JZ Kolter
arXiv preprint arXiv:2011.08105, 2020
652020
Deep abstract q-networks
M Roderick, C Grimm, S Tellex
arXiv preprint arXiv:1710.00459, 2017
392017
Mean actor critic
C Allen, K Asadi, M Roderick, A Mohamed, G Konidaris, M Littman
arXiv preprint arXiv:1709.00503, 2017
34*2017
Provably safe pac-mdp exploration using analogies
M Roderick, V Nagarajan, Z Kolter
International Conference on Artificial Intelligence and Statistics, 1216-1224, 2021
112021
Implementing the deep q-network. arXiv
M Roderick, J MacGlashan, S Tellex
arXiv preprint arXiv:1711.07478, 2017
62017
The AmphibiaWeb app and use of mobile devices in research and outreach
M Roderick, J Gross
Herpetology Notes 7, 109-113, 2014
22014
Generative Posterior Networks for Approximately Bayesian Epistemic Uncertainty Estimation
M Roderick, F Berkenkamp, F Sheikholeslami, Z Kolter
arXiv preprint arXiv:2312.17411, 2023
2023
Projected Off-Policy Q-Learning (POP-QL) for Stabilizing Offline Reinforcement Learning
M Roderick, G Manek, F Berkenkamp, JZ Kolter
arXiv preprint arXiv:2311.14885, 2023
2023
Ensuring the Safety of Reinforcement Learning Algorithms at Training and Deployment
M Roderick
Carnegie Mellon University, 2023
2023
Systems and methods for estimating input certainty for a neural network using generative modeling
M Roderick, F Berkenkamp, F Sheikholeslami, J Kolter
US Patent App. 17/488,096, 2023
2023
Ensuring Safety at Every Stage of the Reinforcement Learning Pipeline
M Roderick
Carnegie Mellon University Pittsburgh, PA, 2022
2022
Controller with neural network and improved stability
JZ Kolter, M Roderick, PL Donti, J Vinogradska
US Patent App. 17/184,995, 2021
2021
Interacting with an unsafe physical environment
D Reeb, JZ Kolter, M Roderick, V Nagarajan
US Patent App. 17/121,237, 2021
2021
2023 Theses by Author
JT BLANE, P CASANOVA, V DWIVEDI, TJ GLAZIER, J LACOMIS, ...
DWIVEDI, VISHAL CMU-S3D-22-110 GLAZIER, Thomas J. CMU-S3D-23-110 LACOMIS, Jeremy CMU-S3D-23-103 MAGELINSKI, Thomas CMU-S3D-23-101
M RODERICK, ZR SHI, J SHIN, W DIVENCENZO, DG WIDDER
El sistema no puede realizar la operación en estos momentos. Inténtalo de nuevo más tarde.
Artículos 1–17