Create Events “accelerate-lunchtime-seminar-series-2”

Author: Kcl36

_events/accelerate-lunchtime-seminar-series-2.md

@@ -0,0 +1,32 @@
+---
+layout: event-single
+title: Accelerate Lunchtime Seminar Series
+start: March 24, 2025 12:00 PM
+end: March 10, 2025 1:00 PM
+image: /assets/uploads/2025-03-24-lunchtime-seminar-1-.jpg
+---
+Join us to find out more about research taking place in AI for Science across the Accelerate Science community.
+
+Details of future talks are available on [Talks@Cam](https://talks.cam.ac.uk/show/index/191074)
+
+Lunch provided, please register to attend via [this form](https://forms.office.com/Pages/ResponsePage.aspx?id=RQSlSfq9eUut41R7TzmG6SCH_8-s-LhNq5ASf8etR39UOTU2TFlIWUw5SkdNNkwyMkI5STdUUlQ4VC4u) so we can confirm catering arrangements. 
+
+**What Connects Us All? From Fibres to Graphs to Neurons**
+
+Akanksha Ahuja, PhD student, Department of Engineering, University of Cambridge
+
+Every time you send a message or watch a video online, your data travels through a vast network of cables spanning cities and continents. Using artificial intelligence and graph theory, we examine the underlying patterns of these backbone networks so we can optimally design, scale and manage them.
+
+In this talk, we’ll explore three questions:
+
+*Can we map the physical shape of the internet? 
+Can we forecast how these networks will expand?
+Can we teach machines to design networks and build the internet of the future?*
+We answer these three questions through the lens of graph analysis, link prediction and graph generation. Our framework transforms network design from a manual process to an intelligent and automated science.
+
+**A Multi-Agent System for Mathematical Discovery**
+Daattavya Aggarwal, Department of Computer Science, University of Cambridge
+
+The practice of working mathematicians shows that the process of discovering novel, interesting mathematics often involves discussions amongst multiple experts and making mistakes. Making false (yet interesting) conjectures and failed attempts to prove them can be driving forces for progress in the field. Moreover, there is an inherent social aspect to judging both the correctness and value of research level math. In this talk, I discuss our proposal for a multi-agent reinforcement learning architecture that incorporates these aspects of the mathematical process, aiming to learn interesting statements purely from mathematical data.
+
+These seminars are open to members of the University of Cambridge. For further details, please email accelerate-science@cst.cam.ac.uk.