The Leverhulme Interdisciplinary Network on Algorithmic Solutions (LINAS) Seminar series is student-led and organised by the doctoral scholars on the programme.

The speakers will be:

Dylan Magill (School of Maths and Physics) 2nd Year LINAS Doctoral Scholar 

Title: 'Preparing for the LSST Tidal Wave: Machine-Learned Simulated Photometry'

LSST is expected to produce a hundredfold increase in the number of transients (such as supernovae) that we discover. However, we are currently not equipped to deal with this deluge of information.  To help prepare for this, I’ve constructed a simulated data set using machine learning grounded in the basis of real observations.   This data will be shared as a classifier challenge and used by others in the community to improve upon their classification algorithms, thereby improving our ability to conduct effective science with LSST.

Dr Ihsen Alouani Senior Lecturer in CSIT, School of EEECS

Title: 'Artificially Intelligent, Inherently Fragile'

As AI models and autonomous agents increasingly influence critical decisions, understanding their vulnerabilities becomes essential for building trust.   This talk explores real-world attack vectors targeting AI systems.   We discuss recent advances in LLM security research that expose fundamental weaknesses in alignment mechanisms.”

We would love it if as many of you as possible could join us for this seminar. Refreshments will also be provided!