GRADSTOP: Early Stopping of Gradient Descent via Posterior Sampling presented at ECAI 2025

HIIT Postdoc Katsiaryna Haitsiukewich and Arash Jamshidi had a privilege to attend the European Conference on Artificial Intelligence (ECAI) 2025 from 25-30 October 25. This year, the conference was held in historical city of Bologna.
Katsiaryna and Arash at ECAI 2025
Katsiaryna Haitsiukevich and Arash Jamshidi at ECAI 2025

The scope of the conference covered core AI topics including machine learning, computer vision, natural language processing, knowledge representation and reasoning, planning and search, as well as multidisciplinary work. Katsiaryna and Arash are delighted that their work was accepted for an oral presentation, and they got a chance to receive feedback from AI community. 

When discussing their research, Katsiaryna stated “we provided a theoretically justified method for early stopping that does not need a validation set, allows using all available data for training, and relies only on the gradients provided by gradient descent algorithm. The method is beneficial in data scarce setting, for example, in processing of medical data."

From the abstract "Machine learning models are often learned by minimising a loss function on the training data using a gradient descent algorithm. These models often suffer from overfitting, leading to a decline in predictive performance on unseen data. A standard solution is early stopping using a hold-out validation set, which halts the minimisation when the validation loss stops decreasing. However, this hold-out set reduces the data available for training. This paper presents GRADSTOP, a novel stochastic early stopping method that only uses information in the gradients, which are produced by the gradient descent algorithm “for free.” Our main contributions are that we estimate the Bayesian posterior by the gradient information, define the early stopping problem as drawing sample from this posterior, and use the approximated posterior to obtain a stopping criterion. Our empirical evaluation shows that GRADSTOP achieves a small loss on test data and compares favourably to a validation-set-based stopping criterion. By leveraging the entire dataset for training, our method is particularly advantageous in data-limited settings, such as transfer learning. It can be incorporated as an optional feature in gradient descent libraries with only a small computational overhead. The source code is available at https://github.com/edahelsinki/gradstop."

Link to the paper: https://ebooks.iospress.nl/doi/10.3233/FAIA251043

More information about ECAI 2025 can be found https://ecai2025.org/

  • Updated:
  • Published:
Share
URL copied!

Read more news

Verena Distler and Raphael Weidhaas
Aalto University, HCI, Highlight, Research Published:

Emotions help explain why people click phishing emails

People are currently taught to recognise phishing attempts in ways that may be counterproductive. Researchers call for information security training to take emotions into account more effectively.
Ja ihminen loi älyn
Aalto University, AI, Highlight Published:

New podcast explores Finnish contributions to artificial intelligence

In Aalto University's podcast “Ja ihminen loi älyn” (And humankind created intelligence), researchers Arno Solin ja Eric Malmi interview pioneers, entrepreneurs and experts about the AI field's history, disruptive consequences and the probability of humanity's destruction.
This AI model follows the logic humans use when choosing where to direct our aattention when reading. Image: Kalle Kataila / Aalto University
AI, Highlight, Research Published:

AI model captures how humans read, paving the way to personalised text and better augmented reality

Researchers now understand not just how our eyes move when we read, but also how we build meaning from text. The new model could hold the key to developing highly customisable applications that adapt to different types of reader.
Kai Puolamäki
AI, Artificial Intelligence, Computer Science Department, Highlight, Interview, University of Helsinki Published:

Professor Kai Puolamäki studies what goes on inside AI

Professor of Computer Science Kai Puolamäki uses AI to investigate new ways of modelling atmospheric particles. In doing so, he wants to open up AI’s black box to enable people to use it with confidence