2023

S. Rezaei-Shoshtari, C. Morissette, F.R. Hogan, G. Dudek, D. Meger. 鈥淗ypernetworks for zero-shot transfer in reinforcement learning鈥�. Proceedings of the AAAI Conference on Artificial Intelligence 37 (8), 9579-9587, 2023.

S. Fujimoto, W.D. Chang, E.J. Smith, S.S. Gu, D. Precup, D. Meger. 鈥淔or SALE: State-Action Representation Learning for Deep Reinforcement Learning鈥�. In Proceedings of the Conference for Neural Information Processing Systems (NeurIPS), 2023.

D. Rivkin, G. Dudek, N. Kakodkar, D. Meger, O. Limoyo, M. Jenkin, X. Liu, F. Hogan. 鈥淎nsel photobot: A robot event photographer with semantic intelligence鈥�. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 8262-8268, 2023.

L. Berry, D. Meger. 鈥淣ormalizing Flow Ensembles for Rich Aleatoric and Epistemic Uncertainty Modeling鈥�. In Proceedings of AAAI 2023.

F. Lotfi, F. Faraji, T. Manderson, D. Meger, G. Dudek. 鈥淐onstrained Robotic Navigation on Preferred Terrains Using Large Language Models and Speech Instruction: Exploiting the Power of Adverbs鈥�. In Proceedings of the 18th International Symposium on Experimental Robotics, 2023.

W.D. Chang, F. Hogan, D. Meger, G. Dudek, 鈥淕eneralizable Imitation Learning Through Pre-Trained Representations鈥�. arXiv preprint arXiv:2311.09350.

W.D. Chang, S. Fujimoto, D. Meger, G. Dudek. 鈥淚mitation Learning from Observation through Optimal Transport鈥�. arXiv preprint arXiv:2310.01632.

F. Lotfi, K. Virji, F. Faraji, L. Berry, A. Holliday, D. Meger, G. Dudek. 鈥淯ncertainty-aware hybrid paradigm of nonlinear MPC and model-based RL for offroad navigation: Exploration of transformers in the predictive model鈥�. arXiv preprint arXiv:2310.00760.

Z. Wang, D. Meger. 鈥淟everaging World Model Disentanglement in Value-Based Multi-Agent Reinforcement Learning鈥�. arXiv preprint arXiv:2309.04615.

L. Berry, D. Meger. 鈥淓scaping the sample trap: Fast and accurate epistemic uncertainty estimation with pairwise-distance estimators. arXiv preprint arXiv:2308.13498.

P. Panangaden, S. Rezaei-Shoshtari, R. Zhao, D. Meger, D. Precup. 鈥淧olicy Gradient Methods in the Presence of Symmetries and State Abstractions鈥�. arXiv preprint arXiv:2305.05666.

L. Berry, D. Meger. 鈥淣ormalizing Flow Ensembles for Rich Aleatoric and Epistemic Uncertainty Modeling鈥�. arXiv preprint arXiv:2302.01312.

2022

Scott Fujimoto et al. 鈥淲hy should i trust you, bellman? the bellman error is a poor replacement for value error鈥�. In: International Conference on Machine Learning. 2022.

Harley E Wiltzer, David Meger, and Marc G Bellemare. 鈥淒istributional Hamiltonjacobi- Bellman equations for continuous-time reinforcement learning鈥�. In: International Conference on Machine Learning. PMLR. 2022, pp. 23832鈥�23856.

Sahand Rezaei-Shoshtari et al. 鈥淐ontinuous MDP Homomorphisms and Homomorphic Policy Gradient鈥�. In: Advances in Neural Information Processing Systems 35 (2022), pp. 20189鈥�20204.

Jonathan Pearce and David Meger. 鈥淎daptive Confidence Calibration鈥�. In: 35th Canadian Conference on Artificial Intelligence. 2022.

Johanna Hansen et al. 鈥淒RIFT-NCRN: A Benchmark Dataset for Drifter Trajectory Prediction鈥�. In: International Conference on Learning Representations, Workshop on AI for Earth and Space Science. 2022.

Edward J Smith et al. 鈥淯ncertainty-Driven Active Vision for Implicit Scene Reconstruction鈥�. In: arXiv preprint arXiv:2210.00978 (2022).

2021

E. Smith, D. Meger, L. Pineda, R. Calandra, J. Malik, A. Romero Soriano, and M. Drozdzal. 鈥淎ctive 3d shape reconstruction from vision and touch,鈥� In Proceedings of the Conference on Advances in Neural Information Processing Systems (NeurIPS), volume 34, 2021.

S. Fujimoto, D. Meger, and D. Precup. 鈥淎 deep reinforcement learning approach to marginalized importance sampling with the successor representation,鈥� In Proceedings of the International Conference on Machine Learning (ICML), 2021.

J.-F. Tremblay, T. Manderson, A. Noca, G. Dudek, and D. Meger. 鈥淢ultimodal dynamics modeling for off-road autonomous vehicles,鈥� In Procedings of the IEEE International Conference on Robotics and Automation (ICRA), 2021.

S. Rezaei-Shoshtari, F. Hogan, M. Jenkin, D. Meger, and G. Dudek. 鈥淟earning intuitive physics with multimodal generative models,鈥� In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI- 21), 2021.

F. Hogan, M. Jenkin, S. Razaei-Shoshtari, Y. Girdhar, D. Meger, and G. Dudek. 鈥淪eeing through your skin: Recognizing objects with a novel visuotactile sensor,鈥� In Proceedings of the Workshop on Applications of Computer Vision (WACV), 2021.

R. Cheng, C.G. Agia, F. Shkurti, D. Meger, and G. Dudek. 鈥淟atent attention augmentation for robust autonomous driving policies,鈥� In Proceedings of the IEEE/RSJ International Conference on Robotics and Intelligent Systems (IROS), 2021.

S. Wapnick, T. Manderson, D. Meger, and G. Dudek. 鈥淭rajectory-constrained deep latent visual attention for improved local planning in presence of heterogeneous terrain,鈥� In Proceedings of the IEEE/ RSJ International Conference on Robotics and Intelligent Systems (IROS), 2021.

D. Rivkin, D. Meger, D. Wu, X. Chen, X. Liu, and G. Dudek. 鈥淟earning assisted identification of scenarios where network optimization algorithms under-perform,鈥� In Proceedings of the IEEE Global Communications Conference (GLOBECOM), 2021.

H. Wang and D. Meger. 鈥淩obotic object manipulation with full-trajectory gan-based imitation learning,鈥� In Proceedings of the Conference on Robotics and Vision (CRV), 2021.

Y. Huang, Y. Yao, J. Hansen, J. Mallette, S. Manjanna, G. Dudek, and D. Meger. 鈥淎n Autonomous Probing System for Collecting Measurements at Depth from Small Surface Vehicles,鈥� In IEEE Oceans Conference and Exposition, 2021.

2020

Edward J Smith, Roberto Calandra, Adriana Romero, Georgia Gkioxari, David Meger, Jitendra Malik, and Michal Drozdzal. 3d shape reconstruction from vision and touch. In Proceedings of the Conference on Neural Information Processing Systems (NeurIPS), 2020.

Scott Fujimoto, David Meger, and Doina Precup. An equivalence between loss functions and non-uniform sampling in experience replay. In Proceedings of the Conference on Neural Information Processing Systems (NeurIPS), 2020.

Travis Manderson, Juan Camilo Gamboa Higuera, Stefan Wapnick, Jean-Francois Tremblay, Florian Shkurti, David Meger, and Gregory Dudek. Vision-based goal-conditioned policies for underwater navigation in the presence of obstacles. In Proceedings of Robotics, Science and Systems (RSS), 2020.

Travis Manderson, Stefan Wapnick, David Meger, and Gregory Dudek. Learning to drive offroad on smooth terrains in unstructured environments using an on-board camera and sparse aerial images. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2020.

Jimmy Li, Karim Koreitem, David Meger, and Gregory Dudek. View-invariant loop closure with oriented semantic landmarks. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2020.

Yi Tian Xu, Xi Chen, Xue Liu, David Meger, and Gregory Dudek. Pressense: Passive respiration sensing via ambient wi signals in noisy environments. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020.

Sahand Rezaei-Shoshtari, David Meger, and Inna Sharf. Learning the latent space of robot dynamics for cutting interaction inference. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020.

Melissa Mozian, Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek. Learning domain randomization distributions for training robust locomotion policies. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020.

Andi Dai and David Meger. Urban night scenery reconstruction by day-night registration and synthesis. In Proceedings of the ACM International Conference on Advances in Geographic Information Systems (SIGSPATIAL), 2020.

Caleb Hoyne, Surya Karthik Mukkavilli, and David Meger. Pmnet: Improving aerosol predictions using deep neural nets for limited ground stations. In Proceedings of the 100th American Meteorological Society Annual Meeting, 2020.

Daniele di Vito, Matthew Bergeron, Gregory Dudek, David Meger, and Gianluca Antonelli. Dynamic planning of redundant robots within a framework of set-based task-priority inverse kinematics. In Proceedings of the IEEE Control Systems Society Conference (CCTA), 2020.

Ran Cheng, David Meger, and Gregory Dudek. Depth prediction for monocular direct visual odometry. In Proceedings of the Confernce on Computer and Robot Vision (CRV), 2020.

Travis Manderson, Juan Camilo Gamboa, Stefan Wapnick, Jean-Francois Tremblay, Hanqing Zhao, Florian Shkurti, David Meger, and Gregory Dudek. Self-supervised, goal-conditioned policies for navigation in unstructured environments. In Proceedings of the Workshop on Workshop on Self-Supervised Robot Learning at Robotics, Science and Systems (RSS) - Best Paper Award, 2020.

Francois Hogan, Sahand Rezaei-Shoshtari, Michael Jenkin, Yogesh Girdhar, David Meger, and Gregory Dudek. Seeing through your skin: A novel visuo-tactile sensor for robotic manipulation. In Proceedings of the Workshop on Visual Learning and Reasoning for Robotic Manipulation at Robotics, Science and Systems (RSS), 2020.

Melissa Mozifan, Amy Zhang, Joelle Pineau, and David Meger. Intervention design for effective sim2real transfer. arXiv preprint arxiv:2012.02055, 2020.

Jean-Francois Tremblay, Travis Manderson, Aurelio Noca, Gregory Dudek, and David Meger. Multimodal dynamics modeling for offroad autonomous vehicles. arXiv preprint arXiv:2011.11751, 2020.

Edward J Smith, Roberto Calandra, Adriana Romero, Georgia Gkioxari, David Meger, Jitendra Malik, and Michal Drozdzal. 3d shape reconstruction from vision and touch. arXiv preprint arXiv:2007.03778, 2020.

Scott Fujimoto, David Meger, and Doina Precup. An equivalence between loss functions and non-uniform sampling in experience replay. arXiv preprint arXiv:2007.06049, 2020.

2019

Edward Smith, Scott Fujimoto, Adriana Romero, and David Meger. 鈥淕eometrics: Exploiting geometric structure for graph-encoded objects鈥�. In Proceedings of the International Conference on Machine Learning (ICML), 2019. Long oral.

Scott Fujimoto, David Meger, and Doina Precup. 鈥淥ff-policy deep reinforcement learning without exploration鈥�. In Proceedings of the International Conference on Machine Learning (ICML), 2019. Short oral and poster.

Jimmy Li, David Meger, and Gregory Dudek. 鈥淪emantic mapping for view-invariant relocalization鈥�. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2019.

Sanjay Thakur, Herke Van Hoof, Juan Camilo Gamboa Higuera, Doina Precup, and David Meger. 鈥淯ncertainty aware learning from demonstrations using bayesian neural networks鈥�. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2019.

Sahand Rezaei-Shoshtari, David Meger, and Sharf Inna. 鈥淐ascaded gaussian processes for data efficient robot dynamics learning鈥�. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019.

Yi Tian Xu, Yaqiao Li, and David Meger. 鈥淗uman motion prediction via pattern completion in latent representation space鈥�. In Proceedings of the Conference on Computer and Robot Vision (CRV), 2019.

Melissa Mozifian, Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek. 鈥淟earning domain randomization distributions for transfer of locomotion policies鈥�. In Proceedings of the Workshop on Multi-Task and Lifelong Reinforcement Learning at ICML, 2019.

Caleb Hoyne, S. Karthik Mukkavilli, and David Meger. 鈥淒eep learning for Aerosol Forecasting鈥�. In Proceedings of the Machine Learning and the Physical Sciences Workshop at the 33rd Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, 2019.

Scott Fujimoto, David Meger and Doina Precup. 鈥淥ff-Policy Deep Reinforcement Learning without Exploration鈥�. Poster presentation at the 4th Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM). Montreal, Canada, 2019.

Scott Fujimoto, David Meger and Doina Precup. 鈥淥ff-Policy Deep Reinforcement Learning without Exploration鈥�. Poster presentation at the Montreal Artificial Intelligence Symposium (MAIS). Montreal, Canada, 2019.

Edward Smith, Scott Fujimoto, Adriana Romero, and David Meger. 鈥淕eometrics: Exploiting geometric structure for graph-encoded objects鈥�. Poster presentation at the Montreal Artificial Intelligence Symposium (MAIS). Montreal, Canada, 2019.

] Xiru Zhu, Fengdi Che, Tianzi Yang, Tzuyang Yu, David Meger, Gregory Dudek. 鈥淒etecting GAN generated errors鈥�. arXiv preprint arXiv:1912.00527, 2019. PAGE 2

Sanjay Thakur, Herke Van Hoof, Gunshi Gupta, David Meger. 鈥淯nifying Variational Inference and PAC-Bayes for Supervised Learning that Scales鈥�. arXiv preprint arXiv:1910.10367, 2019.

2018

Sina Radmard, David Meger, James J. Little, and Elizabeth Croft. Resolving occlusion in active visual target search of high dimensional robotic systems. IEEE Transactions on Robotics (TRO), 34:616 鈥� 629, 2018.

Edward Smith, Scott Fujimoto, David Meger. Multi-View Silhouette and Depth Decomposition for High Resolution 3D Object Representation. In Proceedings of the Thirty-second Conference on Neural Information Processing Systems (NeurIPS), 2018.

Scott Fujimoto, Herke van Hoof, and David Meger. Addressing function approximation error in actor-critic methods. In Proceedings of the International Conference on Machine Learning (ICML), 2018.

Peter Henderson, Wei-Di Chang, Pierre-Luc Bacon, David Meger, Joelle Pineau, and Doina Precup. Optiongan: Learning joint reward-policy options using generative adversarial inverse reinforcement learning. In Proceedings of the Conference of the Association for the Advancement of Artificial Intelligence (AAAI), 2018.

Peter Henderson, Riashat Islam, Philip Bachman, Joelle Pineau, Doina Precup, and David Meger. Deep reinforcement learning that matters. In Proceedings of the Conference of the Association for the Advancement of Artificial Intelligence (AAAI), 2018.

Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek. Synthesizing neural network controllers with probabilistic model-based reinforcement learning. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018.

Peter Henderson, Matthew Vertescher, David Meger, and Mark Coates. Cost adaptation for robust decentralized swarm behaviour. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018.

Victor Barbaros, Herke van Hoof, Abbas Abdolmaleki, and David Meger. Eager and Memory-Based Non-Parametric Stochastic Search Methods for Learning Control. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2018.

Jimmy Li, Zhaoqi Xu, David Meger, and Gregory Dudek. Semantic scene models for visual localization under large viewpoint changes. In Proceedings of the Conference on Computer and Robot Vision (CRV), 2018.

Sanjay Thakur, Herke Van Hoof, Kushal Arora, Doina Precup and David Meger. Sample Efficient Learning From Demonstrations on Multiple Tasks using Bayesian Neural Networks. NeurIPS workshop on Imitation Learning and its Challenges in Robotics (NIPS18-ILR). 2018.

Travis Manderson, Ran Cheng, David Meger and Gregory Dudek. Navigation in the Service of Enhanced Pose Estimation. International Symposium on Experimental Robotics. 2018.

David Meger. Survey of Transfer Learning for Self-Driving. Technical Report submitted to Huawei Canada Research Division under collaborative research agreement. 2018/7/10.

Surya Karthik Mukkavilli, David Meger and Gregory Dudek. EnviRoNet: ImageNet for Environment and Field Robotics. Late breaking abstract at the IEEE International Conference on Robotics and Systems (IROS), 2018.

Scott Fujimoto, David Meger, Doina Precup. Off-Policy Deep Reinforcement Learning without Exploration. arXiv:1812.02900. Publication date 2018/12/7.

2017

Peter Henderson, Wei-Di Chang, Florian Shkurti, Johanna Hansen, David Meger, and Gregory Dudek. Benchmark environments for multitask learning in continuous domains. In The Multitask and Lifelong Learning Workshop at the International Conference on Machine Learning (ICML), 2017.

Peter Henderson, Thang Doan, Riashat Islam, and David Meger. Bayesian policy gradients via alpha divergence dropout inference. In Proceedings of 2nd Bayesian Deep Learning (BDL) Workshop at the Conference on Neural Information Processing Systems (NIPS), 2017.

Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek. Adapting learned robotics behaviours through policy adjustment. In Proceedings of the International Conference on Robotics and Automation (ICRA), 2017.

Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek. From simulation to the field: Learning to swim with the aqua robot. In The Robotic Operating System yearly Conference (ROSCON), September 2017.

Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek. Synthesizing neural network controllers with probabilistic model-based reinforcement learning. In Proceedings of 2nd Bayesian Deep Learning (BDL) Workshop at the Conference on Neural Information Processing Systems (NIPS), 2017.

Jimmy Li, David Meger, and Gregory Dudek. Context-coherent scenes of objects for camera pose estimation. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017.

Edward Smith and David Meger. Improved adversarial systems for 3d object generation and reconstruction. In Proceedings of the Conference on Robot Learning (CoRL), 2017.