Timezone: »
Learning to learn has emerged as an important direction for achieving artificial intelligence. Two of the primary barriers to its adoption are an inability to scale to larger problems and a limited ability to generalize to new tasks. We introduce a learned gradient descent optimizer that generalizes well to new tasks, and which has significantly reduced memory and computation overhead. We achieve this by introducing a novel hierarchical RNN architecture, with minimal per-parameter overhead, augmented with additional architectural features that mirror the known structure of optimization tasks. We also develop a meta-training ensemble of small, diverse, optimization tasks capturing common properties of loss landscapes. The optimizer learns to outperform RMSProp/ADAM on problems in this corpus. More importantly, it performs comparably or better when applied to small convolutional neural networks, despite seeing no neural networks in its meta-training set. Finally, it generalizes to train Inception V3 and ResNet V2 architectures on the ImageNet dataset for thousands of steps, optimization problems that are of a vastly different scale than those it was trained on.
Author Information
Olga Wichrowska (Google Brain)
Niru Maheswaranathan (Stanford University)
Matthew Hoffman (DeepMind)
Sergio Gómez Colmenarejo (Google DeepMind)
Misha Denil (University of Oxford)
Nando de Freitas (DeepMind)
Jascha Sohl-Dickstein (Google Brain)
Related Events (a corresponding poster, oral, or spotlight)
-
2017 Poster: Learned Optimizers that Scale and Generalize »
Tue. Aug 8th 08:30 AM -- 12:00 PM Room Gallery #11
More from the Same Authors
-
2022 Poster: Fast Finite Width Neural Tangent Kernel »
Roman Novak · Jascha Sohl-Dickstein · Samuel Schoenholz -
2022 Spotlight: Fast Finite Width Neural Tangent Kernel »
Roman Novak · Jascha Sohl-Dickstein · Samuel Schoenholz -
2022 Poster: Wide Bayesian neural networks have a simple weight posterior: theory and accelerated sampling »
Jiri Hron · Roman Novak · Jeffrey Pennington · Jascha Sohl-Dickstein -
2022 Spotlight: Wide Bayesian neural networks have a simple weight posterior: theory and accelerated sampling »
Jiri Hron · Roman Novak · Jeffrey Pennington · Jascha Sohl-Dickstein -
2021 Poster: Whitening and Second Order Optimization Both Make Information in the Dataset Unusable During Training, and Can Reduce or Prevent Generalization »
Neha Wadia · Daniel Duckworth · Samuel Schoenholz · Ethan Dyer · Jascha Sohl-Dickstein -
2021 Poster: Unbiased Gradient Estimation in Unrolled Computation Graphs with Persistent Evolution Strategies »
Paul Vicol · Luke Metz · Jascha Sohl-Dickstein -
2021 Spotlight: Whitening and Second Order Optimization Both Make Information in the Dataset Unusable During Training, and Can Reduce or Prevent Generalization »
Neha Wadia · Daniel Duckworth · Samuel Schoenholz · Ethan Dyer · Jascha Sohl-Dickstein -
2021 Oral: Unbiased Gradient Estimation in Unrolled Computation Graphs with Persistent Evolution Strategies »
Paul Vicol · Luke Metz · Jascha Sohl-Dickstein -
2020 Poster: Infinite attention: NNGP and NTK for deep attention networks »
Jiri Hron · Yasaman Bahri · Jascha Sohl-Dickstein · Roman Novak -
2020 Poster: Improving the Gating Mechanism of Recurrent Neural Networks »
Albert Gu · Caglar Gulcehre · Thomas Paine · Matthew Hoffman · Razvan Pascanu -
2019 : Poster discussion »
Roman Novak · Maxime Gabella · Frederic Dreyer · Siavash Golkar · Anh Tong · Irina Higgins · Mirco Milletari · Joe Antognini · Sebastian Goldt · Adín Ramírez Rivera · Roberto Bondesan · Ryo Karakida · Remi Tachet des Combes · Michael Mahoney · Nicholas Walker · Stanislav Fort · Samuel Smith · Rohan Ghosh · Aristide Baratin · Diego Granziol · Stephen Roberts · Dmitry Vetrov · Andrew Wilson · César Laurent · Valentin Thomas · Simon Lacoste-Julien · Dar Gilboa · Daniel Soudry · Anupam Gupta · Anirudh Goyal · Yoshua Bengio · Erich Elsen · Soham De · Stanislaw Jastrzebski · Charles H Martin · Samira Shabanian · Aaron Courville · Shorato Akaho · Lenka Zdeborova · Ethan Dyer · Maurice Weiler · Pim de Haan · Taco Cohen · Max Welling · Ping Luo · zhanglin peng · Nasim Rahaman · Loic Matthey · Danilo J. Rezende · Jaesik Choi · Kyle Cranmer · Lechao Xiao · Jaehoon Lee · Yasaman Bahri · Jeffrey Pennington · Greg Yang · Jiri Hron · Jascha Sohl-Dickstein · Guy Gur-Ari -
2019 : Understanding overparameterized neural networks »
Jascha Sohl-Dickstein -
2019 Poster: Understanding and correcting pathologies in the training of learned optimizers »
Luke Metz · Niru Maheswaranathan · Jeremy Nixon · Daniel Freeman · Jascha Sohl-Dickstein -
2019 Poster: Guided evolutionary strategies: augmenting random search with surrogate gradients »
Niru Maheswaranathan · Luke Metz · George Tucker · Dami Choi · Jascha Sohl-Dickstein -
2019 Oral: Guided evolutionary strategies: augmenting random search with surrogate gradients »
Niru Maheswaranathan · Luke Metz · George Tucker · Dami Choi · Jascha Sohl-Dickstein -
2019 Oral: Understanding and correcting pathologies in the training of learned optimizers »
Luke Metz · Niru Maheswaranathan · Jeremy Nixon · Daniel Freeman · Jascha Sohl-Dickstein -
2019 Poster: Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning »
Natasha Jaques · Angeliki Lazaridou · Edward Hughes · Caglar Gulcehre · Pedro Ortega · DJ Strouse · Joel Z Leibo · Nando de Freitas -
2019 Oral: Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning »
Natasha Jaques · Angeliki Lazaridou · Edward Hughes · Caglar Gulcehre · Pedro Ortega · DJ Strouse · Joel Z Leibo · Nando de Freitas -
2019 Poster: The Effect of Network Width on Stochastic Gradient Descent and Generalization: an Empirical Study »
Daniel Park · Jascha Sohl-Dickstein · Quoc Le · Samuel Smith -
2019 Oral: The Effect of Network Width on Stochastic Gradient Descent and Generalization: an Empirical Study »
Daniel Park · Jascha Sohl-Dickstein · Quoc Le · Samuel Smith -
2018 Poster: Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networks »
Lechao Xiao · Yasaman Bahri · Jascha Sohl-Dickstein · Samuel Schoenholz · Jeffrey Pennington -
2018 Oral: Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networks »
Lechao Xiao · Yasaman Bahri · Jascha Sohl-Dickstein · Samuel Schoenholz · Jeffrey Pennington -
2017 Poster: Learning to Learn without Gradient Descent by Gradient Descent »
Yutian Chen · Matthew Hoffman · Sergio Gómez Colmenarejo · Misha Denil · Timothy Lillicrap · Matthew Botvinick · Nando de Freitas -
2017 Poster: Input Switched Affine Networks: An RNN Architecture Designed for Interpretability »
Jakob Foerster · Justin Gilmer · Jan Chorowski · Jascha Sohl-Dickstein · David Sussillo -
2017 Talk: Input Switched Affine Networks: An RNN Architecture Designed for Interpretability »
Jakob Foerster · Justin Gilmer · Jan Chorowski · Jascha Sohl-Dickstein · David Sussillo -
2017 Poster: Parallel Multiscale Autoregressive Density Estimation »
Scott Reed · Aäron van den Oord · Nal Kalchbrenner · Sergio Gómez Colmenarejo · Ziyu Wang · Yutian Chen · Dan Belov · Nando de Freitas -
2017 Poster: On the Expressive Power of Deep Neural Networks »
Maithra Raghu · Ben Poole · Surya Ganguli · Jon Kleinberg · Jascha Sohl-Dickstein -
2017 Talk: Learning to Learn without Gradient Descent by Gradient Descent »
Yutian Chen · Matthew Hoffman · Sergio Gómez Colmenarejo · Misha Denil · Timothy Lillicrap · Matthew Botvinick · Nando de Freitas -
2017 Talk: On the Expressive Power of Deep Neural Networks »
Maithra Raghu · Ben Poole · Surya Ganguli · Jon Kleinberg · Jascha Sohl-Dickstein -
2017 Talk: Parallel Multiscale Autoregressive Density Estimation »
Scott Reed · Aäron van den Oord · Nal Kalchbrenner · Sergio Gómez Colmenarejo · Ziyu Wang · Yutian Chen · Dan Belov · Nando de Freitas