AI for Materials Discovery in Practice: Effect of low data on RL approaches

Add to your list(s) Download to your calendar using vCal

• Joonatan Laulainen (Altrove)
• Friday 20 March 2026, 15:30-16:30
• Seminar Room 1, Newton Institute.

If you have a question about this talk, please contact nobody.

SCLW02 - Reinforcement Learning for Science: Discovery and Automation

Novel materials have enabled industries and technologies: LEDs, electric motors and batteries. Their development is glacially slow, often taking decades. How do we speed up materials discovery using ML and AI tools. This talk will break down the extreme data challenges of materials discovery. How do we best gain information from 10 or less data points? How do we do that efficiently while accepting that science operates in an exceptionally complex and high-dimensional space? How do we take the correct sequence of decisions that maximise the synthesis outcome for a novel material? How do we bring as much of this complexity ‘in-distribution’? The talk will discuss both traditional ML, LLM -based, and mixed approaches with a holistic view on materials discovery.

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

• All CMS events
• Featured lists
• INI info aggregator
• Isaac Newton Institute Seminar Series
• School of Physical Sciences
• Seminar Room 1, Newton Institute
• bld31
• dh539

Note that ex-directory lists are not shown.