Welcome to the ECML reading group. The aim of the group is to provide a fortnightly meeting where a discussion can take place of an agreed paper in the area of evolutionary computation and/or machine learning. We also aim to hold tutorial sessions on topics of interest to the community.
To lead a paper discussion, present a tutorial, or if you wish to suggest a paper for discussion then please email George De Ath or Tinkle Chugh.
At each reading group the paper lead will briefly give an overview of the paper, either verbally or with slides if they wish, and then lead a discussion of each section of the paper. Note that if a paper is from one of the big machine learning conferences, e.g. NeurIPS (#1, #2), ICLR (#1, #2), and ICML (#1, #2), it is very likely that a recording exists of the paper author’s presentation. These recordings can be used as an alternative to the paper lead presenting an overview of the paper.
Upcoming Sessions
When and Where | Session Details |
---|---|
Date: 11 October 2023 Time: 13:00 - 14:00 Location: Online Presenter: TBA |
|
Date: 25 October 2023 Time: 13:00 - 14:00 Location: Online Presenter: TBA |
Previous Sessions
When and Where | Session Details |
---|---|
Date: 27 September 2023 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Andy Gray, Alma Rahat, Tom Crick, Stephen Lindsay. A Bayesian Active Learning Approach to Comparative Judgement, arXiv, 2023 |
Date: 06 April 2023 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Esteban Real, Chen Liang, David So, Quoc Le. AutoML-Zero: Evolving Machine Learning Algorithms From Scratch, ICML, 2020 |
Date: 08 March 2023 Time: 14:00 - 15:00 Location: Online Presenter: Tinkle |
Jose Pablo Folch, Shiqiang Zhang, Robert M Lee, Behrang Shafei, David Walz, Calvin Tsay, Mark van der Wilk, Ruth Misener. SnAKe: Bayesian Optimization with Pathwise Exploration, NeurIPS, 2022 |
Date: 30 November 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Luke Metz, James Harrison, C. Daniel Freeman, Amil Merchant, Lucas Beyer, James Bradbury, Naman Agrawal, Ben Poole, Igor Mordatch, Adam Roberts, Jascha Sohl-Dickstein. VeLO: Training Versatile Learned Optimizers by Scaling Up, arXiv, 2022 |
Date: 16 November 2022 Time: 14:15 - 15:00 Location: Online Presenter: Marcos |
Dr Marcos Olivera will discuss his ongoing work with a master’s student and present some of their initial results. Its preliminary title is: Explaining the Scaling Limits of Particle Swarm Optimizers. |
Date: 02 November 2022 Time: 13:00 - 14:00 Location: Online Presenter: Ayah Helal |
Johannes Haug, Alexander Braun, Stefan Zürn, Gjergji Kasneci. Change Detection for Local Explainability in Evolving Data Streams, ACM International Conference on Information & Knowledge, 2022 |
Date: 05 October 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Leo Grinsztajn, Edouard Oyallon, Gael Varoquaux. Why do tree-based models still outperform deep learning on tabular data?, NeurIPS, 2022 |
Date: 06 July 2022 Time: 13:00 - 14:00 Location: Online Presenter: Hossein |
Hossein Mohammadi, Peter Challenor. Sequential adaptive design for emulating costly computer codes, arXiv, 2022 |
Date: 08 June 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Yutian Chen, Xingyou Song, Chansoo Lee, Zi Wang, Qiuyi Zhang, David Dohan, Kazuya Kawakami, Greg Kochanski, Arnaud Doucet, Marc’aurelio Ranzato, Sagi Perel, Nando de Freitas. Towards Learning Universal Hyperparameter Optimizers with Transformers, NeurIPS (submission), 2022 |
Date: 25 May 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Aviv Navon, Aviv Shamsian, Ethan Fetaya, Gal Chechik Learning the Pareto Front with Hypernetworks, ICLR, 2021 |
Date: 11 May 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Jiawei Ren, Mingyuan Zhang, Cunjun Yu, Ziwei Liu Balanced MSE for Imbalanced Visual Regression, CVPR, 2022 |
Date: 27 April 2022 Time: 13:00 - 14:00 Location: Online Presenter: Andy |
Kamran Pentland, Massimiliano Tamborrino, T. J. Sullivan, James Buchanan, L. C. Appel GParareal: A time-parallel ODE solver using Gaussian process emulation, British Applied Mathematics Colloquium, 2022 |
Date: 13 April 2022 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Sattar Vakili, Henry Moss, Artem Artemev, Vincent Dutordoir, Victor Picheny Scalable Thompson Sampling using Sparse Gaussian Process Models, NeurIPS, 2021 |
Date: 23 March 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Gowthami Somepalli, Liam Fowl, Arpit Bansal, Ping Yeh-Chiang, Yehuda Dar, Richard Baraniuk, Micah Goldblum, Tom Goldstein Can Neural Nets Learn the Same Model Twice? Investigating Reproducibility and Double Descent from the Decision Boundary Perspective, CVPR, 2022 |
Date: 09 March 2022 Time: 13:00 - 14:00 Location: Online Presenter: Abhra |
Mathilde Caron, Ishan Misra, Julien Mairal, Priya Goyal, Piotr Bojanowski, Armand Jouli. Unsupervised Learning of Visual Features by Contrasting Cluster Assignments, NeurIPS, 2021 |
Date: 23 February 2022 Time: 13:00 - 14:00 Location: Online Presenter: Gregg |
Sebastien Bubeck, Mark Sellke. A Universal Law of Robustness via Isoperimetry, NeurIPS, 2021 |
Date: 09 February 2022 Time: 13:00 - 14:00 Location: Online Presenter: George |
Samuel Müller, Noah Hollmann, Sebastian Pineda Arango, Josif Grabocka, Frank Hutter. Transformers Can Do Bayesian Inference, ICLR, 2022 |
Date: 26 January 2022 Time: 13:00 - 14:00 Location: Online Presenter: Michael |
Michael Dunne will present some of his ongoing work on Variance-Based Sensitivity Analysis for History Matching, which involves conducting Bayesian variance-based sensitivity analysis in regions of not-ruled-out-yet space after History Matching waves, based on the work of Oakley and O’Hagan (2004). For those unfamiliar with History Matching, Wikipedia has a short, but informative, description. |
Date: 15 December 2021 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Sercan O. Arik, Tomas Pfister. TabNet: Attentive Interpretable Tabular Learning, AAAI, 2021 |
Date: 01 December 2021 Time: 13:00 - 14:00 Location: Online Presenter: Andy |
Andrew Corbett, Dmitry Kangin. Imbedding Deep Neural Networks, ICLR (under review), 2021 |
Date: 17 November 2021 Time: 13:00 - 14:00 Location: Online Presenter: Melike |
Gabriela Ochoa, Katherine M. Malan, Christian Blum. Search trajectory networks: A tool for analysing and visualising the behaviour of metaheuristics, Applied Soft Computing, 2021 |
Date: 03 November 2021 Time: 13:00 - 14:00 Location: Online Presenter: George |
Erik Daxberger, Agustinus Kristiadi, Alexander Immer, Runa Eschenhagen, Matthias Bauer, Philipp Hennig. Laplace Redux - Effortless Bayesian Deep Learning, NeurIPS, 2021 |
Date: 06 October 2021 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Jiří Růžička, Jan Koza, Jiří Tumpach, Zbyněk Pitra, Martin Holena. Combining Gaussian Processes with Neural Networks for Active Learning in Optimization, International Workshop on Interactive Adaptive Learning, 2021 |
Date: 22 September 2021 Time: 13:00 - 14:00 Location: Online Presenter: Andy |
Wojciech Samek, Grégoire Montavon, Sebastian Lapuschkin, Christopher J. Anders, Klaus-Robert Müller. Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications, Proceedings of the IEEE, 2021 |
Date: 08 September 2021 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Louis C Tiao, Aaron Klein, Matthias W Seeger, Edwin V. Bonilla, Cedric Archambeau, Fabio Ramos. BORE: Bayesian Optimization by Density-Ratio Estimation, ICML, 2021 |
Date: 28 July 2021 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Brandon Trabucco, Aviral Kumar, Xinyang Geng, Sergey Levine. Conservative Objective Models for Effective Offline Model-Based Optimization, ICML, 2021 |
Date: 30 June 2021 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Henry B. Moss, David S. Leslie, Paul Rayson. BOSH: Bayesian Optimization by Sampling Hierarchically, ICML Workshops, 2020 |
Date: 15 June 2021 Time: 13:00 - 14:00 Location: Online Presenter: Tim |
Lynton Ardizzone, Jakob Kruse, Sebastian Wirkert, Daniel Rahner, Eric Pellegrini, Ralf Klessen, Lena Maier-Hein, Carsten Rother, Ullrich Köthe. Analyzing Inverse Problems with Invertible Neural Networks, ICLR, 2019 |
Date: 19 May 2021 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Jakob Heiss, Jakob Weissteiner, Hanna Wutte, Sven Seuken, Josef Teichmann. NOMU: Neural Optimization-based Model Uncertainty, arXiv, 2021 |
Date: 05 May 2021 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Willie Neiswanger, Ke Alexander Wang, Stefano Ermon. Bayesian Algorithm Execution: Estimating Computable Properties of Black-box Functions Using Mutual Information, ICML, 2021 |
Date: 21 April 2021 Time: 13:00 - 14:00 Location: Online Presenter: Melike |
Hao Li, Zheng Xu, Gavin Taylor, Christoph Studer, Tom Goldstein, Visualizing the Loss Landscape of Neural Nets, Advances in Neural Information Processing Systems 31, 2018 |
Date: 24 March 2021 Time: 13:00 - 14:00 Location: Online Presenter: Multiple |
13:00 Greg Daly Intelligent Process Control – Overcoming the challenge of controlling the processing environment with deep learning, International Conference on Data-Driven Plasma Science, 2021 13:20 Alma Rahat Supporting Policy Decisions in the Covid-19 Pandemic |
Date: 10 March 2021 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Mikhail Belkin, Daniel Hsu, Siyuan Ma, Soumik Mandal, Reconciling modern machine-learning practice and the classical bias–variance trade-off, Proceedings of the National Academy of Sciences, 2019 |
Date: 24 February 2021 Time: 13:00 - 14:00 Location: Online Presenter: Jonathan |
Manuel López-Ibáñez, Joshua Knowles, Marco Laumanns, On Sequential Online Archiving of Objective Vectors, Evolutionary Multi-Criterion Optimization (EMO), 2011 |
Date: 10 February 2021 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Jungtaek Kim, Michael McCourt, Tackgeun You, Saehoon Kim, Seungjin Choi, Bayesian Optimization with Approximate Set Kernels, ECML PKDD, 2021 |
Date: 09 December 2020 Time: 13:00 - 14:00 Location: Online Presenter: George |
Dami Choi, Christopher Shallue, Zachary Nado, Jaehoon Lee, Chris J. Maddison, George E. Dahl On Empirical Comparisons of Optimizers for Deep Learning, arXiv, 2020 |
Date: 25 November 2020 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Yiming Peng, Hisao Ishibuchi, Ke Shang, Multi-modal Multi-objective Optimization: Problem Analysis and Case Studies, IEEE Symposium Series on Computational Intelligence (SSCI), 2019 |
Date: 11 November 2020 Time: 13:00 - 14:00 Location: Online Presenter: George |
Eero Siivola, Akash Kumar Dhaka, Michael Riis Andersen, Javier Gonzalez, Pablo Garcia Moreno, Aki Vehtari, Preferential Batch Bayesian Optimization, arXiv, 2020 |
Date: 28 October 2020 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Xi Lin, Zhiyuan Yang, Qingfu Zhang, Sam Kwong, Controllable Pareto Multi-Task Learning, arXiv, 2020 |
Date: 14 October 2020 Time: 13:00 - 14:00 Location: Online Presenter: George |
Luke Metz, Niru Maheswaranathan, C. Daniel Freeman, Ben Poole, Jascha Sohl-Dickstein, Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves, arXiv, 2020 |
Date: 30 September 2020 Time: 14:00 - 15:00 Location: Online Presenter: Richard |
Novi Quadrianto, Zoubin Ghahramani, A Very Simple Safe-Bayesian Random Forest , IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015 |
Date: 16 September 2020 Time: 13:00 - 14:00 Location: Online Presenter: Melike |
Marcella Scoczynski Ribeiro Martins et al., Multi-layer local optima networks for the analysis of advanced local search-based algorithms , GECCO, 2020 |
Date: 02 September 2020 Time: 13:00 - 14:00 Location: Online Presenter: George |
Vu Nguyen, Michael A Osborne, Knowing The What But Not The Where in Bayesian Optimization , Proceedings of the 37th International Conference on Machine Learning, 2020 |
Date: 19 August 2020 Time: 13:30 - 14:30 Location: Online Presenter: George |
Timur Garipov, Pavel Izmailov, Dmitrii Podoprikhin, Dmitry Vetrov, Andrew Gordon Wilson Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs , Advances in Neural Information Processing Systems 31, 2018 |
Date: 05 August 2020 Time: 13:00 - 14:00 Location: Online Presenter: Fabrizo |
Valerio Perrone, Michele Donini, Krishnaram Kenthapadi, Cédric Archambeau Fair Bayesian Optimization, International Conference on Machine Learning (ICML), 2020 |
Date: 22 July 2020 Time: 13:00 - 14:00 Location: Online Presenter: Mariana |
In this session, Mariana discussed work under review titled: Breast cancer diagnosis using thermal image analysis: an approach based on deep learning and multi-objective binary fish school search for optimized feature selection |
Date: 08 July 2020 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Michael Pearce, Janis Klaise, Matthew Groves, ConBO: Conditional Bayesian Optimization, arXiv, 2020 |
Date: 24 June 2020 Time: 13:00 - 14:30 Location: Online Presenter: Multiple |
Exeter@GECCO2020 13:00 Matt Johns Adaptive Augmented Evolutionary Intelligence for the Design of Water Distribution Networks 13:30 Jonathan Fieldsend Data Structures for Non-Dominated Sets: Implementations and Empirical Assessment of Two Decades of Advances 14:00 Clodomir Joaquim De Santana Junior An Approach to Assess Swarm Intelligence Algorithms Based on Complex Networks |
Date: 10 June 2020 Time: 13:00 - 14:00 Location: Online Presenter: George |
Jack Parker-Holder, Vu Nguyen, & Stephen Roberts, Provably Efficient Online Hyperparameter Optimization with Population-Based Bandits, NeurIPS, 2020 |
Date: 27 May 2020 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Jason Adair, Gabriela Ochoa, and Katherine M. Malan. Local optima networks for continuous fitness landscapes, in Proceedings of the Genetic and Evolutionary Computation Conference Companion (GECCO), 2019 |
Date: 13 May 2020 Time: 13:00 - 14:00 Location: Online Presenter: Richard |
Alexey Dosovitskiy, Josip Djolonga, You Only Train Once: Loss-Conditional Training of Deep Networks, in International Conference on Learning Representations (ICLR), 2020 |
Date: 29 April 2020 Time: 13:00 - 14:00 Location: Online Presenter: George |
Michael Volpp, Lukas P. Fröhlich, Kirsten Fischer, Andreas Doerr, Stefan Falkner, Frank Hutter, Christian Daniel, Meta-Learning Acquisition Functions for Transfer Learning in Bayesian Optimization, in International Conference on Learning Representations (ICLR), 2020 |
Date: 17 April 2020 Time: 13:00 - 14:00 Location: Online Presenter: Tinkle |
Abhishek Gupta, Yew-Soon Ong, Liang Feng, Multifactorial Evolution: Toward Evolutionary Multitasking, in IEEE Transactions on Evolutionary Computation, vol. 20, no. 3, pp. 343-357, 2016 |
Date: 01 April 2020 Time: 15:00 - 16:00 Location: Online Presenter: George |
Stefan Falkner, Aaron Klein, Frank Hutter, BOHB: Robust and Efficient Hyperparameter Optimization at Scale, Proceedings of the 35th International Conference on Machine Learning, in PMLR 80:1437-1446, 2018 |