Prior Specification for Exposure-based Bayesian Matrix Factorization

Zicong Zhu, Issei Sato

Action editor: Seungjin Choi

https://openreview.net/forum?id=o5R4Hv9XqC

#priors #prior #hyperparameters

FoMo-0D: A Foundation Model for Zero-shot Tabular Outlier Detection

Yuchen Shen, Haomin Wen, Leman Akoglu

Action editor: Jiangchao Yao

https://openreview.net/forum?id=XCQzwpR9jE

#outlier #inlier #hyperparameters

New #J2C Certification:

Risk-Controlling Model Selection via Guided Bayesian Optimization

Bracha Laufer-Goldshtein, Adam Fisch, Regina Barzilay, Tommi Jaakkola

https://openreview.net/forum?id=nvmGBcElus

#hyperparameters #guided #optimization

New Study Explains Optimal Settings for Adam Optimizer’s β1 and β2

Researchers find Adam’s β1 = 0.9 and β2 = 0.999 work, but the β1 = √β2 rule is only optimal for certain settings; training benefits from adjusting momentum parameters. getnews.me/new-study-explains-optim... #adamoptimizer #hyperparameters

How far away are truly hyperparameter-free learning algorithms?

Priya Kasimbeg, Vincent Roulet, Naman Agarwal et al.

Action editor: Bryan Kian Hsiang Low

https://openreview.net/forum?id=6BlOCx5c5T

#hyperparameters #hyperparameter #benchmark

Hyperparameters in Continual Learning: A Reality Check

Sungmin Cha, Kyunghyun Cho

Action editor: Elahe Arani

https://openreview.net/forum?id=hiiRCXmbAz

#hyperparameters #hyperparameter #continual

Meta-learning Population-based Methods for Reinforcement Learning

Johannes Hog, Raghu Rajan, André Biedenkapp, Noor Awad, Frank Hutter, Vu Nguyen

Action editor: Mirco Mutti

https://openreview.net/forum?id=d9htascfP8

#bandits #optimize #hyperparameters

Relax and penalize: a new bilevel approach to mixed-binary hyperparameter optimization

Sara Venturini, Marianna De Santis, Jordan Patracone et al.

Action editor: Vlad Niculae

https://openreview.net/forum?id=A1R1cQ93Cb

#hyperparameters #hyperparameter #optimization

The Complete Guide to Support Vector Machines: From Theory to Practice Hyperparameters, advantages, limitations, real-world applications and more

#SVMs are a powerful supervised #ML algorithm for classification & regression. Jamie Crossman-Smith breaks them down with examples, key #hyperparameters, #kernels, and pros/cons. Train & apply them in #KNIME with Leaner-Predictor nodes!

📌 #READ → medium.com/low-code-for...

A thorough reproduction and evaluation of $\mu$P

Georgios Vlassis, David Belius, Volodymyr Fomichov

Action editor: Anastasios Kyrillidis

https://openreview.net/forum?id=AFxEdJwQcp

#hyperparameters #parameters #weights

Transfer Learning in $\ell_1$ Regularized Regression: Hyperparameter Selection Strategy based on ...

Koki Okajima, Tomoyuki Obuchi

Action editor: Bo Han

https://openreview.net/forum?id=ccu0M3nmlF

#lasso #regularized #hyperparameters

An analysis of the noise schedule for score-based generative models

Stanislas Strasman, Antonio Ocello, Claire Boyer, Sylvain Le Corff, Vincent Lemaire

Action editor: Bruno Loureiro

https://openreview.net/forum?id=BlYIPa0Fx1

#generative #wasserstein #hyperparameters

Risk-Controlling Model Selection via Guided Bayesian Optimization

Bracha Laufer-Goldshtein, Adam Fisch, Regina Barzilay, Tommi Jaakkola

Action editor: Pavel Izmailov

https://openreview.net/forum?id=nvmGBcElus

#hyperparameters #guided #optimization

Reading Recommendation: „Beyond algorithm hyperparameters: on preprocessing hyperparameters and associated pitfalls in machine learning applications“ The preprint, which was written by Christina Sauer, Anne-Laure Boulesteix, Luzia Hanßum, Farina Hodiamont, Claudia Bausewein, and Theresa Ullmann, is available on arXiv: Abstract: Adequately genera…

Reading Recommendation: "Beyond algorithm #hyperparameters: on preprocessing hyperparameters and associated pitfalls in machine learning applications" idea.gm.th-koeln.de?p=802

Improving Generalization of Complex Models under Unbounded Loss Using PAC-Bayes Bounds

Xitong Zhang, Avrajit Ghosh, Guangliang Liu, Rongrong Wang

Action editor: Benjamin Guedj

https://openreview.net/forum?id=MP8bmxvWt6

#regularization #priors #hyperparameters

#CausalML update - fitting my first #CausalForest on real data! Does anyone have advice on the most important #hyperparameters? I've got large imbalanced data and a lot of treatment variables, so it's not like anything you see in the economics literature. 🤔 #ML #AI #causal #dataskyence

Scaling Laws Refined: Learning Rate Optimization for Large Language Models Researchers uncovered a scaling law that optimizes learning rates for large language models, enabling better transfer across token horizons and improving training efficiency.

🔍📊🚀 Scaling Laws Refined: Learning Rate Optimization for Large Language Models www.azoai.com/news/2024100... #AI #MachineLearning #LLMs #DeepLearning #ScalingLaws #Optimization #BigData #AIResearch #Hyperparameters #LLama1 @arxiv-stat-ml.bsky.social

📢 Publicationalert: "The Role of Hyperparameters in Machine Learning Models and How to Tune Them" with with Luka Biedebach Andreas Küpfer and Marcel Neunhoeffer in Political Science Research and Methods. Margeret is loving #hyperparameters. Do you? doi.org/10.1017/psrm... 🧵 [1/5]

#publicationalert #Hyperparameters matter for #MachineLearning. In PSRM, @chrisguarnold.bsky.social, Luka Biedebach, Marcel Neunhoeffer, and I show that only 20.31% of top PolSci papers report their HP choices and how they tuned them: doi.org/10.1017/psrm... (1/3)