Workshop
NeurIPS 2022 Workshop on Meta-Learning
Huaxiu Yao 路 Eleni Triantafillou 路 Fabio Ferreira 路 Joaquin Vanschoren 路 Qi Lei
Theater C
Fri 2 Dec, 7 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, improving 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.
Some of the fundamental questions that this workshop aims to address are:
- What are the meta-learning processes in nature (e.g., in humans), and how can we take inspiration from them?
- What is the relationship between meta-learning, continual learning, and transfer learning?
- What interactions exist between meta-learning and large pretrained / foundation models?
- What principles can we learn from meta-learning to help us design the next generation of learning systems?
- What kind of theoretical principles can we develop for meta-learning?
- How can we exploit our domain knowledge to effectively guide the meta-learning process and make it more efficient?
- How can we design better benchmarks for different meta-learning scenarios?
As prospective participants, we primarily target machine learning researchers interested in the questions and foci outlined above. Specific target communities within machine learning include, but are not limited to: meta-learning, AutoML, reinforcement learning, deep learning, optimization, evolutionary computation, and Bayesian optimization. We also invite submissions from researchers who study human learning and neuroscience, to provide a broad and interdisciplinary perspective to the attendees.
Schedule
Fri 7:00 a.m. - 7:10 a.m.
|
Opening remarks
(
Opening remarks
)
>
SlidesLive Video |
馃敆 |
Fri 7:10 a.m. - 7:40 a.m.
|
Invited talk: Mengye Ren
(
Invited talk
)
>
SlidesLive Video |
馃敆 |
Fri 7:40 a.m. - 8:10 a.m.
|
Invited talk: Lucas Beyer
(
Invited talk
)
>
SlidesLive Video |
馃敆 |
Fri 8:10 a.m. - 8:25 a.m.
|
Contributed Talk 1: FiT: Parameter Efficient Few-shot Transfer Learning
(
Contributed Talk
)
>
SlidesLive Video |
馃敆 |
Fri 8:25 a.m. - 8:40 a.m.
|
Break
|
馃敆 |
Fri 8:40 a.m. - 9:40 a.m.
|
Poster session 1 ( poster session ) > link | 馃敆 |
Fri 9:40 a.m. - 9:55 a.m.
|
Contributed talk 2: Optimistic Meta-Gradients
(
contributed talk
)
>
SlidesLive Video |
馃敆 |
Fri 9:55 a.m. - 10:25 a.m.
|
Invited talk: Elena Gribovskaya
(
invited talk
)
>
SlidesLive Video |
馃敆 |
Fri 10:25 a.m. - 12:00 p.m.
|
Lunch break
|
馃敆 |
Fri 12:00 p.m. - 12:30 p.m.
|
Invited talk: Chelsea Finn
(
invited talk
)
>
SlidesLive Video |
馃敆 |
Fri 12:30 p.m. - 1:00 p.m.
|
Invited talk: Greg Yang
(
invited talk
)
>
SlidesLive Video |
馃敆 |
Fri 1:00 p.m. - 1:15 p.m.
|
Contributed talk 3: The Curse of Low Task Diversity: On the Failure of Transfer Learning to Outperform MAML and Their Empirical Equivalence
(
contributed talk
)
>
SlidesLive Video |
馃敆 |
Fri 1:15 p.m. - 2:15 p.m.
|
Poster session 2
(
poster session
)
>
|
馃敆 |
Fri 2:15 p.m. - 2:30 p.m.
|
Contributed talk 4: HyperSound: Generating Implicit Neural Representations of Audio Signals with Hypernetworks
(
contributed talk
)
>
SlidesLive Video |
馃敆 |
Fri 2:30 p.m. - 3:00 p.m.
|
Invited talk: Percy Liang
(
invited talk
)
>
SlidesLive Video |
馃敆 |
Fri 3:00 p.m. - 3:50 p.m.
|
Discussion panel
(
discussion panel
)
>
SlidesLive Video |
馃敆 |
Fri 3:50 p.m. - 4:00 p.m.
|
Closing remarks
(
Closing remarks
)
>
|
馃敆 |
-
|
LOTUS: Learning to learn with Optimal Transport in Unsupervised Scenarios
(
Poster
)
>
link
SlidesLive Video |
prabhant singh 路 Joaquin Vanschoren 馃敆 |
-
|
Test-time adaptation with slot-centric models
(
Poster
)
>
link
SlidesLive Video |
Mihir Prabhudesai 路 Sujoy Paul 路 Sjoerd van Steenkiste 路 Mehdi S. M. Sajjadi 路 Anirudh Goyal 路 Deepak Pathak 路 Katerina Fragkiadaki 路 Gaurav Aggarwal 路 Thomas Kipf 馃敆 |
-
|
Meta-Learning Makes a Better Multimodal Few-shot Learner
(
Poster
)
>
link
SlidesLive Video |
Ivona Najdenkoska 路 Xiantong Zhen 路 Marcel Worring 馃敆 |
-
|
Efficient Bayesian Learning Curve Extrapolation using Prior-Data Fitted Networks
(
Poster
)
>
link
SlidesLive Video |
Steven Adriaensen 路 Herilalaina Rakotoarison 路 Samuel M眉ller 路 Frank Hutter 馃敆 |
-
|
Adversarial Cheap Talk ( Poster ) > link | Chris Lu 路 Timon Willi 路 Alistair Letcher 路 Jakob Foerster 馃敆 |
-
|
Achieving a Better Stability-Plasticity Trade-off via Auxiliary Networks in Continual Learning
(
Poster
)
>
link
SlidesLive Video |
Sanghwan Kim 路 Lorenzo Noci 路 Antonio Orvieto 路 Thomas Hofmann 馃敆 |
-
|
Optimistic Meta-Gradients ( Poster ) > link | Sebastian Flennerhag 路 Tom Zahavy 路 Brendan O'Donoghue 路 Hado van Hasselt 路 Andr谩s Gy枚rgy 路 Satinder Singh 馃敆 |
-
|
Transfer NAS with Meta-learned Bayesian Surrogates
(
Poster
)
>
link
SlidesLive Video |
Gresa Shala 路 Thomas Elsken 路 Frank Hutter 路 Josif Grabocka 馃敆 |
-
|
Gray-Box Gaussian Processes for Automated Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Gresa Shala 路 Andr茅 Biedenkapp 路 Frank Hutter 路 Josif Grabocka 馃敆 |
-
|
AutoRL-Bench 1.0
(
Poster
)
>
link
SlidesLive Video |
Gresa Shala 路 Sebastian Pineda Arango 路 Andr茅 Biedenkapp 路 Frank Hutter 路 Josif Grabocka 馃敆 |
-
|
PersA-FL: Personalized Asynchronous Federated Learning ( Poster ) > link | M. Taha Toghani 路 Soomin Lee 路 Cesar Uribe 馃敆 |
-
|
Bayesian Optimization with a Neural Network Meta-learned on Synthetic Data Only
(
Poster
)
>
link
SlidesLive Video |
Samuel M眉ller 路 Sebastian Pineda Arango 路 Matthias Feurer 路 Josif Grabocka 路 Frank Hutter 馃敆 |
-
|
Recommendation for New Drugs with Limited Prescription Data ( Poster ) > link | Zhenbang Wu 路 Huaxiu Yao 路 Zhe Su 路 David Liebovitz 路 Lucas Glass 路 James Zou 路 Chelsea Finn 路 Jimeng Sun 馃敆 |
-
|
Towards Automated Design of Bayesian Optimization via Exploratory Landscape Analysis
(
Poster
)
>
link
SlidesLive Video |
Carolin Benjamins 路 Anja Jankovic 路 Elena Raponi 路 Koen van der Blom 路 Marius Lindauer 路 Carola Doerr 馃敆 |
-
|
One-Shot Optimal Design for Gaussian Process Analysis of Randomized Experiments
(
Poster
)
>
link
SlidesLive Video |
Jelena Markovic 路 Qing Feng 路 Eytan Bakshy 馃敆 |
-
|
Learning to Prioritize Planning Updates in Model-based Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Brad Burega 路 John Martin 路 Michael Bowling 馃敆 |
-
|
GraViT-E: Gradient-based Vision Transformer Search with Entangled Weights
(
Poster
)
>
link
SlidesLive Video |
Rhea Sukthanker 路 Arjun Krishnakumar 路 sharat patil 路 Frank Hutter 馃敆 |
-
|
Expanding the Deployment Envelope of Behavior Prediction via Adaptive Meta-Learning
(
Poster
)
>
link
SlidesLive Video |
Boris Ivanovic 路 James Harrison 路 Marco Pavone 馃敆 |
-
|
PriorBand: HyperBand + Human Expert Knowledge
(
Poster
)
>
link
SlidesLive Video |
Neeratyoy Mallik 路 Carl Hvarfner 路 Danny Stoll 路 Maciej Janowski 路 Edward Bergman 路 Marius Lindauer 路 Luigi Nardi 路 Frank Hutter 馃敆 |
-
|
The Curse of Low Task Diversity: On the Failure of Transfer Learning to Outperform MAML and Their Empirical Equivalence
(
Poster
)
>
link
SlidesLive Video |
Brando Miranda 路 Patrick Yu 路 Yu-Xiong Wang 路 Sanmi Koyejo 馃敆 |
-
|
Towards Discovering Neural Architectures from Scratch
(
Poster
)
>
link
SlidesLive Video |
Simon Schrodi 路 Danny Stoll 路 Robin Ru 路 Rhea Sukthanker 路 Thomas Brox 路 Frank Hutter 馃敆 |
-
|
HyperSound: Generating Implicit Neural Representations of Audio Signals with Hypernetworks
(
Poster
)
>
link
SlidesLive Video |
Filip Szatkowski 路 Karol J. Piczak 路 Przemys艂aw Spurek 路 Jacek Tabor 路 Tomasz Trzcinski 馃敆 |
-
|
On the Importance of Architectures and Hyperparameters for Fairness in Face Recognition
(
Poster
)
>
link
SlidesLive Video |
Samuel Dooley 路 Rhea Sukthanker 路 John Dickerson 路 Colin White 路 Frank Hutter 路 Micah Goldblum 馃敆 |
-
|
Few-Shot Calibration of Set Predictors via Meta-Learned Cross-Validation-Based Conformal Prediction
(
Poster
)
>
link
SlidesLive Video |
Sangwoo Park 路 Kfir M. Cohen 路 Osvaldo Simeone 馃敆 |
-
|
Multi-objective Tree-structured Parzen Estimator Meets Meta-learning
(
Poster
)
>
link
SlidesLive Video |
Shuhei Watanabe 路 Noor Awad 路 Masaki Onishi 路 Frank Hutter 馃敆 |
-
|
Unsupervised Meta-learning via Few-shot Pseudo-supervised Contrastive Learning
(
Poster
)
>
link
SlidesLive Video |
Huiwon Jang 路 Hankook Lee 路 Jinwoo Shin 馃敆 |
-
|
Uncertainty-Aware Meta-Learning for Multimodal Task Distributions
(
Poster
)
>
link
SlidesLive Video |
Cesar Almecija 路 Apoorva Sharma 路 Young-Jin Park 路 Navid Azizan 馃敆 |
-
|
Lightweight Prompt Learning with General Representation for Rehearsal-free Continual Learning
(
Poster
)
>
link
SlidesLive Video |
Hyunhee Chung 路 Kyung Ho Park 馃敆 |
-
|
Meta-RL for Multi-Agent RL: Learning to Adapt to Evolving Agents
(
Poster
)
>
link
SlidesLive Video |
Matthias Gerstgrasser 路 David Parkes 馃敆 |
-
|
Neural Architecture for Online Ensemble Continual Learning
(
Poster
)
>
link
SlidesLive Video |
Mateusz W贸jcik 路 Witold Ko艣ciukiewicz 路 Adam Gonczarek 路 Tomasz Kajdanowicz 馃敆 |
-
|
Meta-Learning via Classifier(-free) Guidance
(
Poster
)
>
link
SlidesLive Video |
Elvis Nava 路 Seijin Kobayashi 路 Yifei Yin 路 Robert Katzschmann 路 Benjamin F. Grewe 馃敆 |
-
|
MARS: Meta-learning as score matching in the function space
(
Poster
)
>
link
SlidesLive Video |
Kruno Lehman 路 Jonas Rothfuss 路 Andreas Krause 馃敆 |
-
|
Debiasing Meta-Gradient Reinforcement Learning by Learning the Outer Value Function
(
Poster
)
>
link
SlidesLive Video |
Cl茅ment Bonnet 路 Laurence Midgley 路 Alexandre Laterre 馃敆 |
-
|
GramML: Exploring Context-Free Grammars with Model-Free Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Hernan C. Vazquez 路 Jorge Sanchez 路 Rafael Carrascosa 馃敆 |
-
|
Efficient Queries Transformer Neural Processes
(
Poster
)
>
link
SlidesLive Video |
Leo Feng 路 Hossein Hajimirsadeghi 路 Yoshua Bengio 路 Mohamed Osama Ahmed 馃敆 |
-
|
Meta-learning of Black-box Solvers Using Deep Reinforcement Learning
(
Poster
)
>
link
SlidesLive Video |
Cedric Malherbe 路 Aladin Virmaux 路 Ludovic Dos Santos 路 Sofian Chaybouti 馃敆 |
-
|
Contextual Squeeze-and-Excitation
(
Poster
)
>
link
SlidesLive Video |
Massimiliano Patacchiola 路 John Bronskill 路 Aliaksandra Shysheya 路 Katja Hofmann 路 Sebastian Nowozin 路 Richard Turner 馃敆 |
-
|
Conditional Neural Processes for Molecules
(
Poster
)
>
link
SlidesLive Video |
Miguel Garcia-Ortegon 路 Andreas Bender 路 Sergio Bacallado 馃敆 |
-
|
Meta-Learning General-Purpose Learning Algorithms with Transformers
(
Poster
)
>
link
SlidesLive Video |
Louis Kirsch 路 Luke Metz 路 James Harrison 路 Jascha Sohl-Dickstein 馃敆 |
-
|
Betty: An Automatic Differentiation Library for Multilevel Optimization
(
Poster
)
>
link
SlidesLive Video |
Sang Keun Choe 路 Willie Neiswanger 路 Pengtao Xie 路 Eric Xing 馃敆 |
-
|
FiT: Parameter Efficient Few-shot Transfer Learning
(
Poster
)
>
link
SlidesLive Video |
Aliaksandra Shysheya 路 John Bronskill 路 Massimiliano Patacchiola 路 Sebastian Nowozin 路 Richard Turner 馃敆 |
-
|
Topological Continual Learning with Wasserstein Distance and Barycenter
(
Poster
)
>
link
SlidesLive Video |
Tananun Songdechakraiwut 路 Xiaoshuang Yin 路 Barry Van Veen 馃敆 |
-
|
Multiple Modes for Continual Learning ( Poster ) > link | Siddhartha Datta 路 Nigel Shadbolt 馃敆 |
-
|
Interpolating Compressed Parameter Subspaces ( Poster ) > link | Siddhartha Datta 路 Nigel Shadbolt 馃敆 |
-
|
HARRIS: Hybrid Ranking and Regression Forests for Algorithm Selection
(
Poster
)
>
link
SlidesLive Video |
Lukas Fehring 路 Jonas Hanselle 路 Alexander Tornede 馃敆 |