I am a postdoctoral scholar at Stanford University and a recipient of NSF/CRA Computing Innovation Fellowship for research on active learning of transferable priors, kernels, and latent representations for robotics. Currently, I work at the IPRL lab headed by Jeannette Bohg.
I completed my PhD work on data-efficient simulation-to-reality transfer at the Robotics, Perception and Learning lab in KTH (Stockholm, Sweden) working in the group headed by Danica Kragic. My thesis was on "Transfer-Aware Kernels, Priors and Latent Spaces from Simulation to Real Robots". During my PhD time, I also had an opportunity to intern at NVIDIA Robotics (Seattle, USA) and Microsoft Research (Cambridge, UK).
Previously, I was a Masters student at the Robotics Institute at Carnegie Mellon University, developing Bayesian optimization approaches for learning control parameters for bipedal locomotion (with Akshara Rai and Chris Atkeson). During my time at CMU my advisor was Emma Brunskill and in her group I worked on developing Reinforcement Learning algorithms for education.
A few years earlier, I was a software engineer at Google, first in the Search Personalization group and then in the Character Recognition team (developing open-source OCR engine Tesseract).
R Antonova *, J Yang *, K Jatavallabhula, J Bohg. Rethinking Optimization with Differentiable Simulation from a Global Perspective. To appear in Conference on Robot Learning (CoRL), 2022. Selected for oral presentation (6.5% acceptance rate).
R. Antonova, J. Yang, P. Sundaresan, D. Fox, F. Ramos, J. Bohg. A Bayesian Treatment of Real-to-Sim for Deformable Object Manipulation. IEEE Robotics and Automation Letters (RA-L), 2022. [arXiv]
M Li, R Antonova, D Sadigh, J Bohg. Learning Tool Morphology for Contact-Rich Manipulation Tasks with Differentiable Simulation. arXiv:2211.02201, Oct 2022.
C Pan, M Lepert, S Yuan, R Antonova, J Bohg. Task-Driven In-Hand Manipulation of Unknown Objects with Tactile Sensing. arXiv:2210.13403, Oct 2022.
P Sundaresan, R Antonova, J Bohg. DiffCloud: Real-to-Sim from Point Clouds with Differentiable Simulation and Rendering of Deformable Objects. To appear in IEEE International Conference on Intelligent Robots and Systems (IROS), 2022.
J Yang, J Zhang, C Settle, A Rai, R Antonova, J Bohg. Learning Periodic Tasks from Human Demonstrations. IEEE International Conference on Robotics and Automation (ICRA), 2022.
R. Antonova *, P. Shi *, H. Yin, Z. Weng, D. Kragic. Dynamic Environments with Deformable Objects. Neural Information Processing Systems (NeurIPS), Datasets and Benchmarks Track, 2021.
R. Antonova, F. Ramos, R. Possas, D. Fox. BayesSimIG: Scalable Parameter Inference for Adaptive Domain Randomization with IsaacGym, arXiv:2107.04527, 2021. This open source framework was presented as part of the full-day tutorial on End-to-end GPU-accelerated Learning and Control for Robotics with Isaac Gym at RSS, 2021. [Video]
R. Antonova *, A. Varava *, P. Shi, J. Carvalho, D. Kragic. Sequential Topological Representations for Predictive Models of Deformable Objects. In Proceedings of Machine Learning Research as part of the conference on Learning for Dynamics and Control (L4DC), 2021.
R. Antonova, M. Maydanskiy, D. Kragic, S. Devlin, K. Hofmann. Analytic Manifold Learning: Unifying and Evaluating Representations for Continuous Control. Presented at the Conference on Mathematical Theory of Deep Neural Networks (DeepMath), 2020.
M. Hwasser, D. Kragic, R. Antonova. Variational Auto-Regularized Alignment for Sim-to-Real Control. IEEE International Conference on Robotics and Automation (ICRA), 2020.
|ICRA2020 Talk||ICRA2020 Experiments Video|
R. Antonova *, A. Rai *, T. Li, D. Kragic. Bayesian Optimization in Variational Latent Spaces with Dynamic Compression. Conference on Robot Learning (CoRL), PMLR 100: 456-465, 2019
|CoRL2019 Talk||CoRL2019 Experiments Video|
R. Antonova *, M. Kokic *, JA Stork, D. Kragic. Global Search with Bernoulli Alternation Kernel for Task-Oriented Grasping Informed by Simulation. Conference on Robot Learning (CoRL), PMLR 87: 641-650, 2018. Selected for oral presentation (7.5% acceptance rate).
A. Rai *, R. Antonova *, F. Meier, C. Atkeson. Using Simulation to Improve Sample Efficiency of Bayesian Optimization for Bipedal Robots. Journal of Machine Learning Research (JMLR) Bayesian Optimization Issue, (49): 1-24, 2019.
A. Rai *, R. Antonova *, S. Song, W. Martin, H. Geyer, C. Atkeson. Bayesian Optimization Using Domain Knowledge on the ATRIAS Biped. IEEE International Conference on Robotics and Automation (ICRA), 2018 [arXiv, poster]
R. Antonova *, A. Rai *, C. Atkeson. Deep Kernels for Optimizing Locomotion Controllers . Conference on Robot Learning (CoRL), PMLR 78: 47-56, 2017. [arXiv, poster]
R. Antonova *, A. Rai *, C. Atkeson. Sample Efficient Optimization for Learning Controllers for Bipedal Locomotion . IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids), 2016. [arXiv]
|ICRA2018 Talk||CoRL2017 Talk|
R. Antonova *, S. Cruciani *. Unlocking the Potential of Simulators: Design with RL in Mind. Presented at the Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM), 2017. [arXiv, poster]
R. Antonova. Bayesian Optimization for Policy Search: Case Studies in Robotics and Education . Masters thesis. Robotics Institute, Carnegie Mellon University, 2016.
R. Antonova, J. Runde, M. H. Lee, E. Brunskill. Automatically Learning to Teach to the Learning Objectives. In Proceedings of ACM Conference on Learning @ Scale, 2016. [poster]
R. Antonova, J. Runde, C. Dann, E. Brunskill. Improving the Sample Efficiency of Bayesian Optimization Policy Search for Optimal Stopping Problems. Workshop on Data-Efficient Machine Learning at ICML, 2016. [workshop website]
R. Antonova. Multi-task Value of Information Planning for Sequential Multi-task Bandits. Technical Report. Robotics Institute, Carnegie Mellon University, 2016. [related WiML poster]
R. Antonova, A. Handa. Robots Teaching Humans: A New Communication Paradigm via Reverse Teleoperation. International Conference on Autonomous Agents and Multiagent Systems (AAAMAS), 2022.
R. Antonova, J. Bohg. Learning to be Multimodal : Co-evolving Sensory Modalities and Sensor Properties. Conference on Robot Learning, PMLR 164:1782-1788, 2022.
Part of the IEEE Robotics and Automation (RAS) Technical Committee on Robot Learning (junior co-chair).
Part of the organizing committee for Learning for Dynamics and Control (L4DC) 2022 conference.
Program Committee Chair for RSS Pioneers 2021.
One of core organizers for the Physical Reasoning and Inductive Biases for the Real World NeurIPS 2021 workshop.
Co-organizer of Representing and Manipulating Deformable Objects ICRA 2021 workshop.
Co-organizer of a full-day tutorial End-to-end GPU-accelerated Learning and Control for Robotics with Isaac Gym at the Robotics: Science and Systems Conference, 2021.
Carlo Rapisarda. Reinforcement Learning for Dexterity Transfer Between Manipulators. KTH Master's Thesis, June 2019
Martin Hwasser. Converting Deterministic Simulators to Realistic Stochastic Models via Data Alignment. KTH Master's Thesis, June 2019