João Carvalho

João joined the Intelligent Autonomous Systems group as a Ph.D. student in November 2019. He received a M.Sc. degree in Computer Science from the Albert-Ludwigs-Universität Freiburg, and previously completed a Master's degree in Electrical and Computer Engineering from the Instituto Superior Técnico of the University of Lisbon. His master thesis was written at IAS under the supervision of Samuele Tosatto and explored an approach to obtain an off-policy gradient with higher sample efficiency. Currently, he is working within the IKIDA project to develop algorithms that enable robots to work alongside humans.

During his Ph.D. João is developing learning algorithms for robotic manipulation. These include methods that leverage generative models for motion planning, reinforcement learning methods to solve contact-rich tasks like insertions, or improving policy gradient methods with variance reduction techniques.

Publications

Imitation Learning with Deep Generative Models

1. •       Bib
     Funk, N.; Urain, J.; Carvalho, J.; Prasad, V.; Chalvatzaki, G.; Peters, J. (submitted). ActionFlow: Equivariant, Accurate, and Efficient Policies with Spatially Symmetric Flow Matching.

Motion Planning with Diffusion Models (and Optimization)

1. •   Bib
     Le, A. T.; Hansel, K.; Carvalho, J.; Urain, J.; Biess, A.; Chalvatzaki, G.; Peters, J. (2024). Global Tensor Motion Planning, CoRL 2024 Workshop on Differentiable Optimization Everywhere.
2. •       Bib
     Carvalho, J.; Le, A. T.; Baierl, M.; Koert, D.; Peters, J. (2023). Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
3. •     Bib
     Carvalho, J.; Baierl, M; Urain, J; Peters, J. (2022). Conditioned Score-Based Models for Learning Collision-Free Trajectory Generation, NeurIPS 2022 Workshop on Score-Based Methods.

Reactive Motion Planning

1. •     Bib
     Vorndamme, J.; Carvalho, J.; Laha, R.; Koert, D.; Figueredo, L.; Peters, J.; Haddadin, S. (2022). Integrated Bi-Manual Motion Generation and Control shaped for Probabilistic Movement Primitives, 2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids).

Robot Learning for Contact-Rich Manipulation

1. •     Bib
     Carvalho, J.; Koert, D.; Daniv, M.; Peters, J. (2022). Adapting Object-Centric Probabilistic Movement Primitives with Residual Reinforcement Learning, 2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids).

Reinforcement Learning and Policy Gradients

1. •     Bib
     Palenicek, D.; Lutter, M.; Carvalho, J.; Peters, J. (2023). Diminishing Return of Value Expansion Methods in Model-Based Reinforcement Learning, International Conference on Learning Representations (ICLR).
2. •     Bib
     Carvalho, J.; Peters, J. (2022). An Analysis of Measure-Valued Derivatives for Policy Gradients, Multi-disciplinary Conference on Reinforcement Learning and Decision Making (RLDM).
3. •     Bib
     Carvalho, J., Tateo, D., Muratore, F., Peters, J. (2021). An Empirical Analysis of Measure-Valued Derivatives for Policy Gradients, International Joint Conference on Neural Networks (IJCNN).
4. •     Bib
     Tosatto, S.; Carvalho, J.; Peters, J. (2022). Batch Reinforcement Learning with a Nonparametric Off-Policy Policy Gradient, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 44, 10, pp.5996--6010.
5. •     Bib
     Tosatto, S.; Carvalho, J.; Abdulsamad, H.; Peters, J. (2020). A Nonparametric Off-Policy Policy Gradient, Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS).
6. •     Bib
     Carvalho, J.A.C. (2019). Nonparametric Off-Policy Policy Gradient, Master Thesis.