Bristol Machine Learning #18 - Rich Riley

We’re very pleased to announce that our next meet-up will be hosted by Berkeley Square (https://bsqtalent.com) at their Bristol office.

Address: 470 Bath Rd, Bristol BS4 3AP

Please note that spaces for this talk are limited; please only RSVP if you are confident you can attend. Attendance is restricted to those aged 18 and over. Parking is available.

The talk will be recorded and made publicly available after the event.

Schedule

18:00 – Doors open, food, drinks, networking
18:30 – Opening introduction and community messages
18:45 – Rich Riley: Introduction to Vector Symbolic Architectures
19:30 – Food, drinks, networking
19:50 – Close and departure

Talk

Rich Riley: Introduction to Vector Symbolic Architectures

Most of the software we write lives firmly in the world of symbols: variables, data structures, conditionals, and explicit rules. Modern machine learning, by contrast, relies on opaque, distributed representations that are powerful but often difficult to interpret or reason about.

Vector Symbolic Architectures (VSAs) sit between these two traditions. They provide a way to represent symbols, objects, sequences, and even computation itself using high-dimensional vectors, combining the clarity of symbolic systems with the robustness and parallelism of connectionist models.

In this talk, we will build VSAs from first principles. We will see how simple vector operations can encode structured data, tolerate noise, and execute basic programs such as finite state machines.

If you are interested in alternative models of computation, interpretable AI, or how symbolic structure can emerge from high-dimensional geometry, this talk will offer a concrete and intuitive hands-on introduction.

Speaker Bio

Rich Riley is a machine learning engineer and researcher with a broad interest in machine learning and artificial intelligence. His work focuses on applied research, building systems that solve real-world problems while exploring alternative models of representation, learning, and computation.