Papers under review:
An empirical investigation of the challenges of real-world reinforcement learning
G Dulac-Arnold, N Levine, DJ Mankowitz, J Li, C Paduraru, S Gowal, T. Hester
Under review (2020) [PDF]
​
Balancing Constraints and Rewards with Meta-Gradient D4PG
DA Calian*, DJ Mankowitz*, T Zahavy, Z Xu, J Oh, N Levine, T Mann
Under review (2020) [PDF]
​
Robust Constrained Reinforcement Learning for Continuous Control with Model Misspecification
D.J. Mankowitz et. al.
Under review (2020)
​
Publications:​
RL unplugged: Benchmarks for offline reinforcement learning
Caglar Gulcehre, Ziyu Wang, Alexander Novikov, Tom Le Paine, Sergio Gómez Colmenarejo, Konrad Zolna, Rishabh Agarwal, Josh Merel, Daniel Mankowitz, Cosmin Paduraru, Gabriel Dulac-Arnold, Jerry Li, Mohammad Norouzi, Matt Hoffman, Ofir Nachum, George Tucker, Nicolas Heess, Nando de Freitas
(NeurIPS 2020) [PDF]
​
Robust reinforcement learning for continuous control with model misspecification
DJ Mankowitz, N Levine, R Jeong, Y Shi, J Kay, A Abdolmaleki, Y. Shi, T. Hester, M. Riedmiller
(ICLR 2020) [PDF]
​
A bayesian approach to robust reinforcement learning
E Derman, D Mankowitz, T Mann, S Mannor
(UAI 2020) [PDF]
​
Reward Constrained Policy Optimization
C. Tessler, D.J Mankowitz, S. Mannor (ICLR), 2019 [PDF]
​
Universal Successor Feature Approximators
D. Borsa, A. Barreto, J. Quan, D.J Mankowitz, R. Munos, H. Hasselt, D. Silver, T. Schaul, (ICLR), 2019 [PDF]
​
Learn What Not to Learn: Action Elimination with Deep Reinforcement Learning
T. Zahavy, M. Haroush, N. Merlis, D.J Mankowitz, S. Mannor (NIPS), 2018 [PDF]
​
Soft-Robust Actor Critic Policy Gradient
E. Derman, D.J Mankowitz, T.A. Mann, S. Mannor, (UAI), 2018 [PDF]
​
Transfer in deep reinforcement learning using successor features and generalised policy improvement
A. Barreto, D. Borsa, J. Quan, T. Schaul, D. Silver, M. Hessel, D.J Mankowitz, A. Zidek, R. Munos (ICML), 2018 [PDF]
​
Learning Robust Options
D.J Mankowitz, T.A. Mann, P. Bacon, D. Precup, S. Mannor, (AAAI), 2018 [PDF]
​
Shallow Updates for Deep Reinforcement Learning
N. Levine, T. Zahavy, D.J Mankowitz, A. Tamar, S. Mannor, (NIPS), Long Beach, 2017 [PDF]
​
A Deep Hierarchical Approach to Lifelong Learning in Minecraft
C. Tessler*, S. Givony*, T. Zahavy*, D.J. Mankowitz*, S. Mannor, (AAAI), San Francisco, 2017 [PDF] (* - equally contributed)
​
Adaptive Skills, Adaptive Partitions (ASAP)
D.J Mankowitz, T.A. Mann, S. Mannor, To appear In Proc. Neural Information Processing Systems (NIPS), Barcelona, Spain, 2016 [PDF]
​
Time-regularized Interrupting Options
D.J Mankowitz, T.A. Mann, S. Mannor, In Proc. International Conference on Machine Learning (ICML), Beijing, China, 2014 [PDF]
​
BRISK-based Visual Feature Extraction for Resource Constrained Robots
D.J. Mankowitz, S. Ramamoorthy, In Proc. RoboCup International Symposium, Eindhoven, Netherlands, 2013 [PDF]
​
Mobile Device-Based Cellular Network Coverage Analysis Using Crowd Sourcing
D.J. Mankowitz, A. Paverd, IEEE Xplore, EUROCON International Conference on the Computer as a Tool, Lisbon, Portugal, 2011 [PDF]​
Patents
2019: Continual reinforcement learning with a multi-task agent
T Schaul, M Hessel, HP Van Hasselt, DJ Mankowitz
US Patent App. 16/268,414
​
2018: Systems and Methods for Promoting Products in Product Search Results Using Transfer Learning with Active Sampling, Publication number: 20180211303, Publication Date: July 26, 2018
R.E. Chatwin, D.J Mankowitz, S. Mannor, V. Abhishek
​
Workshops/Technical Papers
​
Local Search for Policy Iteration in Continuous ControlJost Tobias Springenberg, Nicolas Heess, Daniel Mankowitz, Josh Merel, Arunkumar Byravan, Abbas Abdolmaleki, Jackie Kay, Jonas Degrave, Julian Schrittwieser, Yuval Tassa, Jonas Buchli, Dan Belov, Martin Riedmiller
Arxvi 2020 [PDF]
​
Action assembly: Sparse imitation learning for text based games with combinatorial action spaces
C Tessler, T Zahavy, D Cohen, DJ Mankowitz, S Mannor
RLDM 2019 [PDF]
​
Challenges of real-world reinforcement learning
G Dulac-Arnold, D Mankowitz, T Hester
ICML 2019 - RL4RealLife Workshop (Best paper award) [PDF]
​
Bootstrapping Skills
D.J Mankowitz, T.A. Mann, S. Mannor, In Proc. Reinforcement Learning and Decision Making (RLDM), Edmonton, Canada, 2015 (Selected for Oral Presentation <15% Acceptance)
​
Iterative Hierarchical Optimization for Misspecified Problems (IHOMP)
D.J Mankowitz, T.A. Mann, S. Mannor, European Conference on Reinforcement Learning (EWRL), Barcelona, 2016
​
Situational Awareness by Risk-Conscious Skills
D.J Mankowitz, A. Tamar, S. Mannor, International Conference on Machine Learning (ICML) Workshop: Machine Learning in the Wild, New York, 2016 [PDF] (selected for Oral Presentation)
​
Learning when to Switch Between Skills in High Dimensional Domains
T.A. Mann, D.J Mankowitz, S. Mannor, Workshop on Learning for General Competency in Video Games, AAAI, Austin, Texas, 2015 [PDF]
​
A Deep Hierarchical Approach to Lifelong Learning in Minecraft
C. Tessler, S. Givony, T. Zahavy, D.J. Mankowitz, S. Mannor, IJCAI Deep Learning Workshop, New York, 2016 [PDF] and EWRL 2016 (selected for Oral Presentation) [PDF]
​
CFORB: Circular FREAK ORB Visual Odometry
D.J. Mankowitz, E. Rivlin, Technical Report, Technion, Israel Institute of Technology, 2014 [ArXiv preprint]
​
A Review and Comparison of the Various Techniques used for Online Simultaneous Localisation and Mapping (SLAM)
D.J. Mankowitz, Technical Report, University of Edinburgh, Scotland, 2011
​
Theses:
Towards General AI: Learning Options in Hierarchical Reinforcement Learning
D.J. Mankowitz, Technion Israel Institutre of Technology, PhD Thesis, 2018
​
BRISK-based Visual Landmark Localisation using Nao Humanoid Robots
D.J. Mankowitz, MSc Thesis, University of Edinburgh, Scotland, 2012
​
Mobile Device-Based Cellular Network Coverage Analysis Using Crowd Sourcing
D.J. Mankowitz, BEng Thesis, University of the Witwatersrand, Johannesburg, South Africa, 2010
​
Co-authors:
-
Andre Barreto
-
Chen Tessler
-
Matteo Hessel
-
John Quan
-
Junhyuk Oh
-
Dan Horgan
-
Esther Derman