AI, ML and Friends is a weekly seminar series within the School of Computing on Artificial Intelligence, Machine Learning, and related topics. We are open to attendees and presenters external to the school. Please sign up to the mailing list to receive weekly announcements including zoom details, and email the seminar organiser to schedule a talk.
Upcoming Seminars
10 November 2026, 10:30#
Uncertainty quantification in the age of foundation models#
Speaker: Dr Russell Tsuchida
Abstract: Uncertainty quantification — an important consideration in almost all fields of science, social science, and engineering — allows practitioners to make risk-aware decisions, know when to trust a prediction, improve explainability, and keep humans in the loop. Recently, foundation models and AI have promised to fundamentally change the landscape of uncertainty quantification research. In this talk, I will describe an evolving research program in uncertainty quantification, spanning work before and after the rise of foundation models. I will first present published work on squared families (the topic of my DECRA), sequential decision making, and Bregman divergences, with applications in point cloud alignment and label distribution learning. Bregman divergences form a natural mathematical language for evaluating and training uncertainty-aware models, including foundation models. Building on this, I will then look ahead to two directions: using tabular foundation models for uncertainty quantification, and using Bregman divergences to advance diffusion and related generative models.This seminar will include a mix of published results and a longer-term research vision for uncertainty quantification in the age of foundation models.
Bio: Dr Russell Tsuchida is a 2026 ARC DECRA Fellow and continuing academic at Monash University’s Faculty of Information Technology. He completed his doctorate in probabilistic machine learning at The University of Queensland in 2020. He then joined Data61, first as a CSIRO Early Research Career Fellow, and from 2023, as a research scientist. Dr Tsuchida’s work advances our understanding of core probabilistic machine learning methods such as Bayesian neural networks, Gaussian processes, and density modelling. He publishes regularly at leading venues including NeurIPS, ICML, ICLR, and AAAI, and his work has received a NeurIPS spotlight and a TMLR Expert Certification. He is an Action Editor for TMLR and an Area Chair for ICML, and has received reviewer awards from NeurIPS, ICLR, ICML, AISTATS, and TMLR.
Where: Building 145, room 1.33
19 November 2026, 11:00#
Evaluating Conversational Agents at Scale#
Speaker: Paul Thomas
Abstract: Microsoft 365 Copilot helps millions of people around the world navigate their digital world and get things done. Measuring and improving Copilot’s quality relies on constant and in-depth evaluations, but these evaluations are made difficult by precisely the same features that make the product interesting: it is multimodal, multilingual, multiturn, serves a huge number and diversity of people, and has strong guarantees on data protection and privacy. I’ll outline some of the challenges of evaluating such a system at scale, and some of the techniques we’ve developed to help understand and improve the product.
Bio: Paul Thomas is an applied scientist at Microsoft, where he has worked on Bing and Copilot. Both his applied work and his research looks at how people use information access systems, especially search systems and conversational systems, and how we should evaluate these. He’s involved in the Conference on Human Information Interaction and Retrieval (CHIIR), the Text Retrieval Conference (TREC), and SIGIR, and has worked with students at RMIT University and the ANU. Before Microsoft, Paul worked at the CSIRO and the ANU, especially on distributed search, NLP, and evaluation.
Where: zoom
26 November 2026, 11:00#
To score or not to score#
Speaker: Chen Soon Ong
Abstract: Scoring the performance of a machine learning model is a key choice we all make, whether it is the loss for training a model, a gap for understanding alignment, or a metric such as accuracy for predictive performance. This talk considers the score we use in adaptive experimental design for scientific discovery, where we use bandit type approaches to choose experiments. I show some recent theoretical and practical work on adaptive design and discuss the needs of cross disciplinary collaboration for AI in science, for example to empirically validate bandit designs in a genomics lab. On the other side, machine learning researchers may be tempted to generalise the idea of optimising a score to other aspects of work and life. I argue that it is problematic to convert the diversity of human experience into a single scalar score, and then optimising that score. Finally I discuss some preliminary results on what could be in between scoring and not scoring.
Bio: Cheng Soon Ong is the science director at the Technology Research Unit in CSIRO. He works on extending machine learning methods and enabling new approaches to scientific discovery, and has led programs of research including foundational machine learning in science and purpose driven machine learning applications at CSIRO. He supervises and mentors many junior scientists, collaborates with scientists on problems in genomics and astronomy, advocates for open science, and is an adjunct Associate Professor at the Australian National University. He is co-author of the textbook Mathematics for Machine Learning, and his career has spanned multiple roles in Malaysia, Germany, Switzerland, and Australia.
Where: TBD
03 December 2026, 11:00#
###
Speaker: tbd
Abstract:
Where: Building 145, room 1.33
10 December 2026, 11:00#
###
Speaker: Pushkar Gopalakrishna
Abstract: Autonomous LLM agents can become trapped in repetitive execution loops, excessive tool calls, or recursive delegation while continuing to consume computational resources. Detecting these failures is challenging because legitimate long-running tasks can exhibit similar behavior. This talk presents an experimental investigation of resource governance and runaway-behavior detection in autonomous agents, drawing on approximately 16,500 simulated agent runs and offline analysis of over 2,200 real-agent trajectories from publicly available benchmarks. The presentation examines hierarchical budget enforcement, loop detection, circuit breakers, and the trade-offs between preventing runaway execution and preserving legitimate agent workflows. It also discusses failure cases where improvements to legitimate-task completion significantly increased the cost of stopping runaway agents, highlighting broader challenges in reliable agent evaluation.
Bio: Pushkar Gopalakrishna is a Staff Software Engineer at Snap Inc., with over 16 years of experience building large-scale distributed systems and cloud infrastructure. Previously, he held engineering roles at Amazon and Cruise. His research interests include autonomous AI reliability, distributed systems, cloud-native computing, and resource optimization. He has authored research papers accepted at IEEE conferences and has participated in technical speaking engagements and academic peer-review activities.
Where: zoom
15 December 2026, 10:30#
###
Speaker: MoonJeong and Saemi
Abstract:
Where: Building 145, room 1.33
17 December 2026, 11:00#
###
Speaker: Alexander Soen
Abstract:
Where: Building 145, room 1.33