AI Agents: From Perception to Autonomous Action
29 September 2026
19 attendees
Description
From models that predict to agents that reason, plan, and act.
Speakers and topics
AI Agents: From Perception to Autonomous Action
Steven Wallace, PhD Researcher at the University of Aberdeen
Recent advances in foundation models have given rise to AI agents capable of reasoning, planning, and acting autonomously across complex tasks. This talk explored the shift from static, single-purpose deep learning models towards agentic systems that combine perception, multi-modal understanding, and decision-making to operate independently in real-world environments. Drawing on examples from natural language processing and beyond, we examined the opportunities of building AI agents that are capable of carrying out complex multi-step tasks.
Speaker bios
Steven Wallace
Steven Wallace is a PhD researcher at the University of Aberdeen working on deep learning foundation models for semantic segmentation in climate change analysis. With a background spanning energy consultancy, electronic engineering, and AI, he has built systems ranging from explainable heart disease classifiers to real-time surveillance object detection, and has taught and guest-lectured on deep learning and image segmentation at Aberdeen. Beyond computer vision, his research interests extend to multi-modal learning and AI agents.
Slides
Event Photos
