Workshop
5th Workshop on Meta-Learning
Erin Grant 路 F谩bio Ferreira 路 Frank Hutter 路 Jonathan Richard Schwarz 路 Joaquin Vanschoren 路 Huaxiu Yao
Mon 13 Dec, 3 a.m. PST
Recent years have seen rapid progress in meta-learning methods, which transfer knowledge across tasks and domains to efficiently learn new tasks, optimize the learning process itself, and even generate new learning methods from scratch. Meta-learning can be seen as the logical conclusion of the arc that machine learning has undergone in the last decade, from learning classifiers, to learning representations, and finally to learning algorithms that themselves acquire representations, classifiers, and policies for acting in environments. In practice, meta-learning has been shown to yield new state-of-the-art automated machine learning methods, novel deep learning architectures, and substantially improved one-shot learning systems. Moreover, to improve one鈥檚 own learning capabilities through experience can also be viewed as a hallmark of intelligent beings, and neuroscience shows a strong connection between human and reward learning and the growing sub-field of meta-reinforcement learning.
Schedule
Mon 3:00 a.m. - 3:10 a.m.
|
Introduction and opening remarks
(
Live speech
)
>
SlidesLive Video |
馃敆 |
Mon 3:10 a.m. - 3:35 a.m.
|
Ying Wei
(
Invited talk
)
>
SlidesLive Video |
Ying Wei 馃敆 |
Mon 3:35 a.m. - 3:40 a.m.
|
Ying Wei Q&A ( Q&A ) > link | Ying Wei 馃敆 |
Mon 3:40 a.m. - 4:00 a.m.
|
Meta-Learning Reliable Priors in the Function Space
(
Contributed talk & Poster
)
>
link
SlidesLive Video |
Jonas Rothfuss 路 Dominique Heyn 路 jinfan Chen 路 Andreas Krause 馃敆 |
Mon 4:00 a.m. - 5:00 a.m.
|
Poster session 1 ( Poster session ) > link | 馃敆 |
Mon 5:00 a.m. - 5:25 a.m.
|
Carlo Ciliberto
(
Invited talk
)
>
SlidesLive Video |
Carlo Ciliberto 馃敆 |
Mon 5:25 a.m. - 5:30 a.m.
|
Carlo Ciliberto Q&A ( Q&A ) > link | Carlo Ciliberto 馃敆 |
Mon 5:30 a.m. - 5:55 a.m.
|
Mihaela Van Der Schaar
(
Invited talk
)
>
SlidesLive Video |
Mihaela van der Schaar 馃敆 |
Mon 5:55 a.m. - 6:00 a.m.
|
Mihaela Van Der Schaar Q&A ( Q&A ) > link | Mihaela van der Schaar 馃敆 |
Mon 6:00 a.m. - 7:00 a.m.
|
Break
|
馃敆 |
Mon 7:00 a.m. - 8:00 a.m.
|
Panel Discussion
(
Panel Discussion
)
>
link
SlidesLive Video |
馃敆 |
Mon 8:00 a.m. - 8:20 a.m.
|
Bootstrapped Meta-Learning
(
Contributed talk & Poster
)
>
link
SlidesLive Video |
Sebastian Flennerhag 路 Yannick Schroecker 路 Tom Zahavy 路 Hado van Hasselt 路 David Silver 路 Satinder Singh 馃敆 |
Mon 8:20 a.m. - 8:45 a.m.
|
Nan Rosemary Ke
(
Invited talk
)
>
SlidesLive Video |
Nan Rosemary Ke 馃敆 |
Mon 8:45 a.m. - 8:50 a.m.
|
Nan Rosemary Ke Q&A ( Q&A ) > link | Nan Rosemary Ke 馃敆 |
Mon 8:50 a.m. - 10:00 a.m.
|
Poster session 2 ( Poster session ) > link | 馃敆 |
Mon 10:00 a.m. - 10:25 a.m.
|
Luke Metz
(
Invited talk
)
>
SlidesLive Video |
Luke Metz 馃敆 |
Mon 10:25 a.m. - 10:30 a.m.
|
Luke Metz Q&A ( Q&A ) > link | Luke Metz 馃敆 |
Mon 10:30 a.m. - 10:55 a.m.
|
Eleni Triantafillou
(
Invited talk
)
>
SlidesLive Video |
Eleni Triantafillou 馃敆 |
Mon 10:55 a.m. - 11:00 a.m.
|
Eleni Triantafillou Q&A ( Q&A ) > link | Eleni Triantafillou 馃敆 |
Mon 11:00 a.m. - 11:20 a.m.
|
Offline Meta-Reinforcement Learning with Online Self-Supervision
(
Contributed talk & Poster
)
>
link
SlidesLive Video |
Vitchyr Pong 路 Ashvin Nair 路 Laura Smith 路 Catherine Huang 路 Sergey Levine 馃敆 |
Mon 11:20 a.m. - 12:30 p.m.
|
Poster session 3 ( Poster session ) > link | 馃敆 |
-
|
Variational Task Encoders for Model-Agnostic Meta-Learning
(
Poster
)
>
link
SlidesLive Video |
Joaquin Vanschoren 馃敆 |
-
|
Meta-learning from sparse recovery
(
Poster
)
>
link
SlidesLive Video |
Beicheng Lou 路 Nathan Zhao 路 Jiahui Wang 馃敆 |
-
|
Open-Ended Learning Strategies for Learning Complex Locomotion Skills
(
Poster
)
>
link
SlidesLive Video |
Joaquin Vanschoren 馃敆 |
-
|
Neural Processes with Stochastic Attention: Paying more attention to the context dataset ( Poster ) > link | Mingyu Kim 路 KyeongRyeol Go 路 Se-Young Yun 馃敆 |
-
|
Transformers Can Do Bayesian-Inference By Meta-Learning on Prior-Data
(
Poster
)
>
link
SlidesLive Video |
Samuel M眉ller 路 Noah Hollmann 路 Sebastian Pineda Arango 路 Josif Grabocka 路 Frank Hutter 馃敆 |
-
|
On the Practical Consistency of Meta-Reinforcement Learning Algorithms
(
Poster
)
>
link
SlidesLive Video |
Zheng Xiong 路 Luisa Zintgraf 路 Jacob Beck 路 Risto Vuorio 路 Shimon Whiteson 馃敆 |
-
|
Understanding Catastrophic Forgetting and Remembering in Continual Learning with Optimal Relevance Mapping
(
Poster
)
>
link
SlidesLive Video |
prakhar kaushik 路 Adam Kortylewski 路 Alex Gain 路 Alan Yuille 馃敆 |
-
|
Few Shot Image Generation via Implicit Autoencoding of Support Sets
(
Poster
)
>
link
SlidesLive Video |
Shenyang Huang 路 Kuan-Chieh Wang 路 Guillaume Rabusseau 路 Alireza Makhzani 馃敆 |
-
|
Meta-learning inductive biases of learning systems with Gaussian processes ( Poster ) > link | Michael Li 路 Erin Grant 路 Tom Griffiths 馃敆 |
-
|
DARTS without a Validation Set: Optimizing the Marginal Likelihood ( Poster ) > link | Miroslav Fil 路 Robin Ru 路 Clare Lyle 路 Yarin Gal 馃敆 |
-
|
Curriculum Meta-Learning for Few-shot Classification ( Poster ) > link | Priyanka Agrawal 馃敆 |
-
|
Transfer Learning for Bayesian HPO with End-to-End Landmark Meta-Features
(
Poster
)
>
link
SlidesLive Video |
Hadi Jomaa 路 Sebastian Pineda Arango 路 Lars Schmidt-Thieme 路 Josif Grabocka 馃敆 |
-
|
Successor Feature Neural Episodic Control
(
Poster
)
>
link
SlidesLive Video |
David Emukpere 路 Xavier Alameda-Pineda 路 Chris Reinke 馃敆 |
-
|
FedMix: A Simple and Communication-Efficient Alternative to Local Methods in Federated Learning
(
Poster
)
>
link
SlidesLive Video |
Elnur Gasanov 路 Ahmed Khaled Ragab Bayoumi 路 Samuel Horv谩th 路 Peter Richtarik 馃敆 |
-
|
Studying BatchNorm Learning Rate Decay on Meta-Learning Inner-Loop Adaptation
(
Poster
)
>
link
SlidesLive Video |
Alexander Wang 路 Sasha (Alexandre) Doubov 路 Gary Leung 馃敆 |
-
|
Effect of diversity in Meta-Learning
(
Poster
)
>
link
SlidesLive Video |
Ramnath Kumar 路 Tristan Deleu 路 Yoshua Bengio 馃敆 |
-
|
How to distribute data across tasks for meta-learning?
(
Poster
)
>
link
SlidesLive Video |
Alexandru Cioba 路 Michael Bromberg 路 Qian Wang 路 RITWIK NIYOGI 路 Georgios Batzolis 路 Jezabel Garcia 路 Da-shan Shiu 路 Alberto Bernacchia 馃敆 |
-
|
Unsupervised Meta-Learning via Latent Space Energy-based Model of Symbol Vector Coupling ( Poster ) > link | Bo Pang 路 Deqian Kong 路 Ying Nian Wu 馃敆 |
-
|
A Nested Bi-level Optimization Framework for Robust Few Shot Learning ( Poster ) > link | Krishnateja Killamsetty 路 Changbin Li 路 Chen Zhao 路 Rishabh Iyer 路 Feng Chen 馃敆 |
-
|
Introducing Symmetries to Black Box Meta Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Louis Kirsch 路 Sebastian Flennerhag 路 Hado van Hasselt 路 Abram Friesen 路 Junhyuk Oh 路 Yutian Chen 馃敆 |
-
|
A Meta-Gradient Approach to Learning Cooperative Multi-Agent Communication Topology ( Poster ) > link | Qi Zhang 路 Dingyang Chen 馃敆 |
-
|
Task Attended Meta-Learning for Few-Shot Learning
(
Poster
)
>
link
SlidesLive Video |
AROOF AIMEN 路 Bharat Ladrecha 路 Narayanan C Krishnan 馃敆 |
-
|
One Step at a Time: Pros and Cons of Multi-Step Meta-Gradient Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Cl茅ment Bonnet 路 Paul Caron 路 Thomas D Barrett 路 Ian Davies 路 Alexandre Laterre 馃敆 |
-
|
Skill-based Meta-Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Taewook Nam 路 Shao-Hua Sun 路 Karl Pertsch 路 Sung Ju Hwang 路 Joseph Lim 馃敆 |
-
|
Efficient Automated Online Experimentation with Multi-Fidelity ( Poster ) > link | Steven Kleinegesse 路 Zhenwen Dai 路 Andreas Damianou 路 Kamil Ciosek 路 Federico Tomasi 馃敆 |
-
|
Sign-MAML: Efficient Model-Agnostic Meta-Learning by SignSGD
(
Poster
)
>
link
SlidesLive Video |
Chen Fan 路 Parikshit Ram 路 Sijia Liu 馃敆 |
-
|
A Preliminary Study on the Feature Representations of Transfer Learning and Gradient-Based Meta-Learning Techniques
(
Poster
)
>
link
SlidesLive Video |
Mike Huisman 路 Jan van Rijn 路 Aske Plaat 馃敆 |
-
|
On the Role of Pre-training for Meta Few-Shot Learning
(
Poster
)
>
link
SlidesLive Video |
Chia-You Chen 路 Hsuan-Tien Lin 路 Masashi Sugiyama 路 Gang Niu 馃敆 |
-
|
Contrastive Embedding of Structured Space for Bayesian Optimization
(
Poster
)
>
link
SlidesLive Video |
Josh Tingey 路 Ciar谩n Lee 路 Zhenwen Dai 馃敆 |
-
|
Hierarchical Few-Shot Generative Models ( Poster ) > link | Giorgio Giannone 路 Ole Winther 馃敆 |