ICLR 2020元学习论文汇总

1.Title	Meta-Q-Learning
Author	Rasool Fakoor, Pratik Chaudhari, Stefano Soatto, Alexander J. Smola
Highlight	MQL is a simple off-policy meta-RL algorithm that recycles data from the meta-training replay buffer to adapt to new tasks
2.Title	Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Tasks
Author	Donghyun Na, Hae Beom Lee, Hayeon Lee, Saehoon Kim, Minseop Park, Eunho Yang, Sung Ju Hwang
Highlight	A novel meta-learning model that adaptively balances the effect of the meta-learning and task-specific learning, and also class-specific learning within each task.
3.Title	Meta-Learning with Warped Gradient Descent
Author	Sebastian Flennerhag, Andrei A. Rusu, Razvan Pascanu, Francesco Visin, Hujun Yin, Raia Hadsell
Highlight	We propose a novel framework for meta-learning a gradient-based update rule that scales to beyond few-shot learning and is applicable to any form of learning, including continual learning.
4.Title	Meta-Learning without Memorization
Author	Mingzhang Yin, George Tucker, Mingyuan Zhou, Sergey Levine, Chelsea Finn
Highlight	We identify and formalize the memorization problem in meta-learning and solve this problem with novel metaregularization method, which greatly expand the domain that meta-learning can be applicable to and effective on.
5.Title	Meta-Learning Acquisition Functions for Transfer Learning in Bayesian Optimization
Author	Michael Volpp, Lukas Froehlich, Kirsten Fischer, Andreas Doerr, Stefan Falkner, Frank Hutter, Christian Daniel
Highlight	We perform efficient and flexible transfer learning in the framework of Bayesian optimization through metalearned neural acquisition functions.
6.Title	Improving Generalization in Meta Reinforcement Learning using Neural Objectives
Author	Louis Kirsch, Sjoerd van Steenkiste, Juergen Schmidhuber
Highlight	We introduce MetaGenRL, a novel meta reinforcement learning algorithm. Unlike prior work, MetaGenRL can generalize to new environments that are entirely different from those used for meta-training
7.Title	Automated Relational Meta-learning
Author	Huaxiu Yao, Xian Wu, Zhiqiang Tao, Yaliang Li, Bolin Ding, Zhenhui Li
Highlight	Addressing task heterogeneity problem in meta-learning by introducing meta-knowledge graph
8.Title	ES-MAML: Simple Hessian-Free Meta Learning
Author	Xingyou Song, Wenbo Gao, Yuxiang Yang, Krzysztof Choromanski, Aldo Pacchiano, Yunhao Tang
Highlight	We provide a new framework for MAML in the ES/blackbox setting, and show that it allows deterministic and linear policies, better exploration, and non-differentiable adaptation operators
9.Title	Watch, Try, Learn: Meta-Learning from Demonstrations and Rewards
Author	Allan Zhou, Eric Jang, Daniel Kappler, Alex Herzog, Mohi Khansari, Paul Wohlhart, Yunfei Bai, Mrinal Kalakrishnan, Sergey Levine, Chelsea Finn
Highlight	We introduce the 2-simplicial Transformer and show that this architecture is a useful inductive bias for logical reasoning in the context of deep reinforcement learning.
10.Title	Few-shot Text Classification with Distributional Signatures
Author	Yujia Bao, Menghua Wu, Shiyu Chang, Regina Barzilay
Highlight	Meta-learning methods used for vision, directly applied to NLP, perform worse than nearest neighbors on new classes; we can do better with distributional signatures.
11.Title	A Theoretical Analysis of the Number of Shots in Few-Shot Learning
Author	Tianshi Cao, Marc T Law, Sanja Fidler
Highlight	The paper analyzes the effect of shot number on prototypical networks and proposes a robust method when the shot number differs from meta-training to meta-testing time.
12.Title	Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML
Author	Aniruddh Raghu, Maithra Raghu, Samy Bengio, Oriol Vinyals
Highlight	The success of MAML relies on feature reuse from the meta-initialization, which also yields a natural simplification of the algorithm, with the inner loop removed for the network body, as well as other insights on the head and body.
13.Title	Differentially Private Meta-Learning
Author	Jeffrey Li, Mikhail Khodak, Sebastian Caldas, Ameet Talwalkar
Highlight	A novel approach using mode connectivity in loss landscapes to mitigate adversarial effects, repair tampered models and evaluate adversarial robustness
14.Title	MetaPix: Few-Shot Video Retargeting
Author	Jessica Lee, Deva Ramanan, Rohit Girdhar
Highlight	Video retargeting typically requires large amount of target data to be effective, which may not always be available; we propose a metalearning approach that improves over popular baselines while producing temporally coherent frames.
15.Title	Towards Fast Adaptation of Neural Architectures with Meta Learning
Author	Dongze Lian, Yin Zheng, Yintao Xu, Yanxiong Lu, Leyu Lin, Peilin Zhao, Junzhou Huang, Shenghua Gao
Highlight	A novel Bayesian deep learning framework that captures and relates hierarchical semantic and visual concepts, performing well on a variety of image and text modeling and generation tasks.
16.Title	Bayesian Meta Sampling for Fast Uncertainty Adaptation
Author	Zhenyi Wang, Yang Zhao, Ping Yu, Ruiyi Zhang, Changyou Chen
Highlight	We proposed a Bayesian meta sampling method for adapting the model uncertainty in meta learning
17.Title	Meta-Learning Deep Energy-Based Memory Models
Author	Sergey Bartunov, Jack Rae, Simon Osindero, Timothy Lillicrap
Highlight	Deep associative memory models using arbitrary neural networks as a storage.
18.Title	Meta-learning curiosity algorithms
Author	Ferran Alet*, Martin F. Schneider*, Tomas Lozano-Perez, Leslie Pack Kaelbling
Highlight	Meta-learning curiosity algorithms by searching through a rich space of programs yields novel mechanisms that generalize across very different reinforcement-learning domains
19.Title	VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning
Author	Luisa Zintgraf, Kyriacos Shiarlis, Maximilian Igl, Sebastian Schulze, Yarin Gal, Katja Hofmann, Shimon Whiteson
Highlight	VariBAD opens a path to tractable approximate Bayes-optimal exploration for deep RL using ideas from metalearning, Bayesian RL, and approximate variational inference.
20.Title	Empirical Bayes Transductive Meta-Learning with Synthetic Gradients
Author	Xu Hu, Pablo Moreno, Yang Xiao, Xi Shen, Guillaume Obozinski, Neil Lawrence
Highlight	We present a learning rule for feedback weights in a spiking neural network that addresses the weight transport problem.
21.Title	Meta Reinforcement Learning with Autonomous Inference of Subtask Dependencies
Author	Sungryull Sohn, Hyunjae Woo, Jongwook Choi, Honglak Lee
Highlight	A novel meta-RL method that infers latent subtask structure
22.Title	A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
Author	Yoshua Bengio, Tristan Deleu, Nasim Rahaman, Nan Rosemary Ke, Sebastien Lachapelle, Olexa Bilaniuk, Anirudh Goyal, Christopher Pal
Highlight	This paper proposes a meta-learning objective based on speed of adaptation to transfer distributions to discover a modular decomposition and causal variables.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

这些论文里还能看到MAML的影子，不得不说Chelsea Finn是真的厉害。。。