Category: Artificial Intelligence

Creative Applications of Machine Learning

speaker: Luba Elliott

Over the past couple of years, there has been increasing interest in applying the latest advances in machine learning to creative projects in art, music, film, theatre and beyond. From Google’s DeepDream and style transfer to the world’s first computer-generated musical playing in London’s West End, more and more creative AI projects are moving beyond the world of research and...

How Deep Learning is changing the car insurance industry

speaker: Giovanni Lotti

Artificial intelligence has been a very popular expression for the last 10 years, but disruptive commercial applications of it are not as common. Tractable is taking a very pragmatic approach to AI and has been working to solve real commercial problems with deep learning technology. Car insurance is a deeply flawed industry, and Tractable is disrupting its problems through deep...

A working introduction to Symbolic AI

speaker: Martin Lambert

Historically there have been two main AI research tracks – symbolic AI (logic) and sub-symbolic AI (machine learning, neural networks, etc.) – but the recent resurgence of interest in AI is focused almost exclusively on machine learning. Why? Human intelligence combines conscious logical reasoning with subconscious pattern recognition. Are we missing even bigger AI opportunities by overlooking symbolic AI? Internet-scale...

Functional data science and algebraic infrastructure

speaker: Nick Pollard

Functional programming concepts such as composition, immutability, and type-safety allow software developers to rapidly build reliable, reusable, and correct systems. These approaches can also be used to solve many of the problems that data scientists face in the wild — such as sharing and reusing models, connecting them to data sources and third-party systems, and ensuring predictable behaviour in production....

Learning representations for unordered item sets

speaker: Andrew Clegg

RNNs and their more sophisticated cousins (GRUs, LSTMs) have proven to be “unreasonably effective” at learning from sequence data, although they can be tricky and expensive to train. But what if your data consists of unordered sets or bags of objects, or the data is ordered but the predictive value of that ordering is marginal? Deep Averaging Networks (Iyyer et...

The mechanism of thought from Peter McElwaine Johnn

Peter is a technologist with over 30 years experience of information systems design and development acquired in both consulting and industry roles across several sectors. He has worked for Deloitte, Microsoft, Wipro and Lloyds Banking Group and specialises in strategy and architecture. His areas of interest and expertise include: the profound impacts from AI and all of the other technologies...