Seguir
Petra Poklukar
Petra Poklukar
Research Scientist, Google Zurich
Email confirmado em google.com
Título
Citado por
Citado por
Ano
Latent space roadmap for visual action planning of deformable and rigid object manipulation
M Lippi, P Poklukar, MC Welle, A Varava, H Yin, A Marino, D Kragic
2020 IEEE/RSJ International Conference on Intelligent Robots and Systems …, 2020
652020
Geometric Multimodal Contrastive Representation Learning
P Poklukar, M Vasco, H Yin, FS Melo, A Paiva, D Kragic
International Conference on Machine Learning, 17782-17800, 2022
632022
Bayesian Meta-Learning for Few-Shot Policy Adaptation Across Robotic Platforms
A Ghadirzadeh, X Chen, P Poklukar, C Finn, M Björkman, D Kragic
2021 IEEE/RSJ International Conference on Intelligent Robots and Systems …, 2021
342021
Brave: Broadening the visual encoding of vision-language models
OF Kar, A Tonioni, P Poklukar, A Kulshrestha, A Zamir, F Tombari
European Conference on Computer Vision, 113-132, 2024
292024
Enabling visual action planning for object manipulation through latent space roadmap
M Lippi, P Poklukar, MC Welle, A Varava, H Yin, A Marino, D Kragic
IEEE Transactions on Robotics 39 (1), 57-75, 2022
182022
Data-efficient visuomotor policy training using reinforcement learning and generative models
A Ghadirzadeh, P Poklukar, V Kyrki, D Kragic, M Björkman
arXiv preprint arXiv:2007.13134, 2020
142020
Delaunay Component Analysis for Evaluation of Data Representations
DK Petra Poklukar, Vladislav Polianskii, Anastasia Varava, Florian Pokorny
International Conference on Learning Representations, 2022
13*2022
GeomCA: Geometric Evaluation of Data Representations
P Poklukar, A Varava, D Kragic
Proceedings of the 38th International Conference on Machine Learning 139 …, 2021
122021
Bayesian meta-learning for few-shot policy adaptation across robotic platforms. In 2021 IEEE
A Ghadirzadeh, X Chen, P Poklukar, C Finn, M Björkman, D Kragic
RSJ International Conference on Intelligent Robots and Systems (IROS), 1274-1280, 0
8
Latent space roadmap for visual action planning of deformable and rigid object manipulation. In 2020 IEEE
M Lippi, P Poklukar, MC Welle, A Varava, H Yin, A Marino, D Kragic
RSJ International Conference on Intelligent Robots and Systems (IROS), 5619-5626, 0
7
Augment-Connect-Explore: a Paradigm for Visual Action Planning with Data Scarcity
M Lippi, MC Welle, P Poklukar, A Marino, D Kragic
2022 IEEE/RSJ International Conference on Intelligent Robots and Systems …, 2022
42022
Modeling assumptions and evaluation schemes: On the assessment of deep latent variable models.
J Bütepage, P Poklukar, D Kragic
CVPR Workshops, 9-12, 2019
42019
GraphDCA--a Framework for Node Distribution Comparison in Real and Synthetic Graphs
C Ceylan, P Poklukar, H Hultin, A Kravchenko, A Varava, D Kragic
arXiv preprint arXiv:2202.03884, 2022
32022
Batch Curation for Unsupervised Contrastive Representation Learning
MC Welle, P Poklukar, D Kragic
arXiv preprint arXiv:2108.08643, 2021
32021
Hyperbolic Delaunay Geometric Alignment
AA Medbouhi, GL Marchetti, V Polianskii, A Kravberg, P Poklukar, ...
Joint European Conference on Machine Learning and Knowledge Discovery in …, 2024
22024
Training and evaluation of deep policies using reinforcement learning and generative models
A Ghadirzadeh, P Poklukar, K Arndt, C Finn, V Kyrki, D Kragic, ...
Journal of machine learning research 23 (174), 1-37, 2022
22022
Few-Shot Learning with Weak Supervision
A Ghadirzadeh, P Poklukar, X Chen, H Yao, H Azizpour, M Björkman, ...
Learning to Learn-Workshop at ICLR 2021, 2021
22021
Seeing the whole picture instead of a single point: Self-supervised likelihood learning for deep generative models
P Poklukar, J Bütepage, D Kragic
Second Symposium on Advances in Approximate Bayesian Inference, 0
1
Hyperbolic Delaunay Geometric Alignment
A Aiman Medbouhi, GL Marchetti, V Polianskii, A Kravberg, P Poklukar, ...
arXiv e-prints, arXiv: 2404.08608, 2024
2024
Learning and Evaluating the Geometric Structure of Representation Spaces
P Poklukar
KTH Royal Institute of Technology, 2022
2022
O sistema não pode efectuar a operação agora. Tente mais tarde.
Artigos 1–20