LQG 30/06/20 – On-Line – How and why Causal AI techniques can prevent overfitting – Darko Matovski

How and why Causal AI techniques can prevent over-fitting 

Darko Matovski – CEO causaLens

In this seminar Darko Matovski will discuss Causal AI a nascent science which aims to enable machines to understand cause and effect. Luminaries such as Judea Pearl and Yoshua Bengio believe that Causal AI is the fix for the weaknesses of current state-of-the-art machine learning, which relies on past patterns and correlations to make predictions of the future.  The current state-of-the-art can work for static environments and closed loop problems with fixed rules, but it often fails in other situations. The likelihood of failure is especially pronounced for highly dynamic, low signal-to-noise environments, such as financial markets. 

 Many prominent academics and AI industry leaders see Causal AI and machine understanding of cause and effect as the route to more accurate and consistent forecasts. An understanding of the true causal drivers should enable causal AI to navigate, draw inferences from and make forecasts for complex, dynamic and metamorphosing systems. Further, it should be capable of ‘imagining’ scenarios not encountered in the past so allowing it to simulate counterfactual worlds to learn from, instead of relying solely on training data.

Understanding causality should also give AI the ability to interact with humans more profoundly, by being able to both incorporate human domain knowledge and explain its ‘thought process’. The academic and industrial AI communities are racing to advance the science of causality and apply its power to solve a wide range of problems.  

Darko will describe and discuss Causal AI and present examples of the areas where it should help machines and humans – especially in finance!

Darko Matovski

CEO causaLens

Dr. Darko Matovski is the CEO of causaLens. The company is leading Causal AI research, a way for machines to understand cause & effect.  Darko has worked for cutting edge hedge funds and research institutions including the National Physical Laboratory in London (where Alan Turing worked) and Man Group in London. Darko has a PhD in Machine Learning and an MBA.