Understanding and correcting pathologies in the training of learned optimizers

24 October 2018

Jascha Narain Sohl-Dickstein

ODL

ArXiv PDF HTML

Papers citing "Understanding and correcting pathologies in the training of learned optimizers"

50 / 51 papers shown

Title
Optimization Problem Solving Can Transition to Evolutionary Agentic Workflows Wenhao Li Bo Jin Mingyi Hong Changhong Lu Xiangfeng Wang 48 0 0 07 May 2025
Discovering Quality-Diversity Algorithms via Meta-Black-Box Optimization Maxence Faldor Robert Tjarko Lange Antoine Cully 81 0 0 04 Feb 2025
Learning Versatile Optimizers on a Compute Diet A. Moudgil Boris Knyazev Guillaume Lajoie Eugene Belilovsky 168 0 0 22 Jan 2025
Can Learned Optimization Make Reinforcement Learning Less Difficult? Alexander David Goldie Chris Xiaoxuan Lu Matthew Jackson Shimon Whiteson Jakob N. Foerster 42 3 0 09 Jul 2024
From Learning to Optimize to Learning Optimization Algorithms Camille Castera Peter Ochs 65 1 0 28 May 2024
$The boundary of neural network trainability is fractal$ The boundary of neural network trainability is fractal Jascha Narain Sohl-Dickstein 36 8 0 09 Feb 2024
Investigation into the Training Dynamics of Learned Optimizers Jan Sobotka Petr Simánek Daniel Vasata 28 0 0 12 Dec 2023
Variance-Reduced Gradient Estimation via Noise-Reuse in Online Evolution Strategies Oscar Li James Harrison Jascha Narain Sohl-Dickstein Virginia Smith Luke Metz 51 5 0 21 Apr 2023
Learning To Optimize Quantum Neural Network Without Gradients Ankit Kulshrestha Xiaoyuan Liu Hayato Ushijima-Mwesigwa Ilya Safro 25 5 0 15 Apr 2023
Symbolic Discovery of Optimization Algorithms Xiangning Chen Chen Liang Da Huang Esteban Real Kaiyuan Wang ... Xuanyi Dong Thang Luong Cho-Jui Hsieh Yifeng Lu Quoc V. Le 67 352 0 13 Feb 2023
Learning to Optimize for Reinforcement Learning Qingfeng Lan Rupam Mahmood Shuicheng Yan Zhongwen Xu OffRL 26 6 0 03 Feb 2023
Mnemosyne: Learning to Train Transformers with Transformers Deepali Jain K. Choromanski Kumar Avinava Dubey Sumeet Singh Vikas Sindhwani Tingnan Zhang Jie Tan OffRL 39 9 0 02 Feb 2023
A Survey of Meta-Reinforcement Learning Jacob Beck Risto Vuorio E. Liu Zheng Xiong L. Zintgraf Chelsea Finn Shimon Whiteson OOD OffRL 37 122 0 19 Jan 2023
Federated Automatic Differentiation Keith Rush Zachary B. Charles Zachary Garrett FedML 34 1 0 18 Jan 2023
Data Distillation: A Survey Noveen Sachdeva Julian McAuley DD 45 73 0 11 Jan 2023
General-Purpose In-Context Learning by Meta-Learning Transformers Louis Kirsch James Harrison Jascha Narain Sohl-Dickstein Luke Metz 40 72 0 08 Dec 2022
Transformer-Based Learned Optimization Erik Gartner Luke Metz Mykhaylo Andriluka C. Freeman C. Sminchisescu 18 11 0 02 Dec 2022
Discovering Evolution Strategies via Meta-Black-Box Optimization R. T. Lange Tom Schaul Yutian Chen Tom Zahavy Valenti Dallibard Chris Xiaoxuan Lu Satinder Singh Sebastian Flennerhag 44 47 0 21 Nov 2022
VeLO: Training Versatile Learned Optimizers by Scaling Up Luke Metz James Harrison C. Freeman Amil Merchant Lucas Beyer ... Naman Agrawal Ben Poole Igor Mordatch Adam Roberts Jascha Narain Sohl-Dickstein 35 60 0 17 Nov 2022
Learning to Learn with Generative Models of Neural Network Checkpoints William S. Peebles Ilija Radosavovic Tim Brooks Alexei A. Efros Jitendra Malik UQCV 75 65 0 26 Sep 2022
An Investigation of the Bias-Variance Tradeoff in Meta-Gradients Risto Vuorio Jacob Beck Shimon Whiteson Jakob N. Foerster Gregory Farquhar 27 8 0 22 Sep 2022
A Closer Look at Learned Optimization: Stability, Robustness, and Inductive Biases James Harrison Luke Metz Jascha Narain Sohl-Dickstein 47 22 0 22 Sep 2022
Gradient-based Bi-level Optimization for Deep Learning: A Survey Can Chen Xiangshan Chen Chen Ma Zixuan Liu Xue Liu 94 35 0 24 Jul 2022
Dataset Distillation using Neural Feature Regression Yongchao Zhou E. Nezhadarya Jimmy Ba DD FedML 47 151 0 01 Jun 2022
Automated Dynamic Algorithm Configuration Steven Adriaensen André Biedenkapp Gresa Shala Noor H. Awad Theresa Eimer Marius Lindauer Frank Hutter 32 36 0 27 May 2022
Meta-AF: Meta-Learning for Adaptive Filters Jonah Casebeer Nicholas J. Bryan Paris Smaragdis 159 28 0 25 Apr 2022
Practical tradeoffs between memory, compute, and performance in learned optimizers Luke Metz C. Freeman James Harrison Niru Maheswaranathan Jascha Narain Sohl-Dickstein 38 32 0 22 Mar 2022
Symbolic Learning to Optimize: Towards Interpretability and Scalability Wenqing Zheng Tianlong Chen Ting-Kuei Hu Zhangyang Wang 45 19 0 13 Mar 2022
Do Differentiable Simulators Give Better Policy Gradients? H. Suh Max Simchowitz Kaipeng Zhang Russ Tedrake 32 95 0 02 Feb 2022
Tutorial on amortized optimization Brandon Amos OffRL 75 43 0 01 Feb 2022
Unbiased Gradient Estimation in Unrolled Computation Graphs with Persistent Evolution Strategies Paul Vicol Luke Metz Jascha Narain Sohl-Dickstein 27 67 0 27 Dec 2021
ModelPred: A Framework for Predicting Trained Model from Training Data Yingyan Zeng Jiachen T. Wang Si-An Chen H. Just Ran Jin R. Jia TDI MU 33 2 0 24 Nov 2021
Gradients are Not All You Need Luke Metz C. Freeman S. Schoenholz Tal Kachman 28 93 0 10 Nov 2021
Learned Robust PCA: A Scalable Deep Unfolding Approach for High-Dimensional Outlier Detection HanQin Cai Jialin Liu W. Yin 35 39 0 11 Oct 2021
Auto-DSP: Learning to Optimize Acoustic Echo Cancellers Jonah Casebeer Nicholas J. Bryan Paris Smaragdis 25 10 0 08 Oct 2021
Introducing Symmetries to Black Box Meta Reinforcement Learning Louis Kirsch Sebastian Flennerhag Hado van Hasselt A. Friesen Junhyuk Oh Yutian Chen 22 30 0 22 Sep 2021
Bootstrapped Meta-Learning Sebastian Flennerhag Yannick Schroecker Tom Zahavy Hado van Hasselt David Silver Satinder Singh 38 59 0 09 Sep 2021
Accelerating Quadratic Optimization with Reinforcement Learning Jeffrey Ichnowski Paras Jain Bartolomeo Stellato G. Banjac Michael Luo Francesco Borrelli Joseph E. Gonzalez Ion Stoica Ken Goldberg OffRL 19 35 0 22 Jul 2021
Learn2Hop: Learned Optimization on Rough Landscapes Amil Merchant Luke Metz S. Schoenholz E. D. Cubuk 31 16 0 20 Jul 2021
Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulation C. Freeman Erik Frey Anton Raichuk Sertan Girgin Igor Mordatch Olivier Bachem 48 350 0 24 Jun 2021
Stateless Neural Meta-Learning using Second-Order Gradients Mike Huisman Aske Plaat Jan N. van Rijn 26 7 0 21 Apr 2021
Learning to Optimize: A Primer and A Benchmark Tianlong Chen Xiaohan Chen Wuyang Chen Howard Heaton Jialin Liu Zhangyang Wang W. Yin 40 225 0 23 Mar 2021
Training Learned Optimizers with Randomly Initialized Learned Optimizers Luke Metz C. Freeman Niru Maheswaranathan Jascha Narain Sohl-Dickstein 43 12 0 14 Jan 2021
Reverse engineering learned optimizers reveals known and novel mechanisms Niru Maheswaranathan David Sussillo Luke Metz Ruoxi Sun Jascha Narain Sohl-Dickstein 22 21 0 04 Nov 2020
Training Stronger Baselines for Learning to Optimize Tianlong Chen Weiyi Zhang Jingyang Zhou Shiyu Chang Sijia Liu Lisa Amini Zhangyang Wang OffRL 25 51 0 18 Oct 2020
Learning the Step-size Policy for the Limited-Memory Broyden-Fletcher-Goldfarb-Shanno Algorithm Lucas N. Egidio A. Hansson B. Wahlberg 16 12 0 03 Oct 2020
Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves Luke Metz Niru Maheswaranathan C. Freeman Ben Poole Jascha Narain Sohl-Dickstein 33 62 0 23 Sep 2020
Meta-Learning in Neural Networks: A Survey Timothy M. Hospedales Antreas Antoniou P. Micaelli Amos Storkey OOD 55 1,935 0 11 Apr 2020
Using learned optimizers to make models robust to input noise Luke Metz Niru Maheswaranathan Jonathon Shlens Jascha Narain Sohl-Dickstein E. D. Cubuk VLM OOD 17 26 0 08 Jun 2019
Bilevel Programming for Hyperparameter Optimization and Meta-Learning Luca Franceschi P. Frasconi Saverio Salzo Riccardo Grazzi Massimiliano Pontil 110 717 0 13 Jun 2018