Currently Senior Research Advisor at Capital Fund Management (CFM, Paris) and visiting researcher at Imperial College (London), Charles-Albert Lehalle is a leading expert in market microstructure and optimal trading. Formerly Global Head of Quantitative Research at Crédit Agricole Cheuvreux, and Head of Quantitative Research on Market Microstructure in the Equity Brokerage and Derivative Department of Crédit Agricole Corporate Investment Bank, he studied intensively the market microstructure evolution since the financial crisis and regulatory changes in Europe and in the US. He provided research and expertise on this topic to investors and intermediaries, and is often heard by regulators and policy-makers like the European Commission, the French Senate, the UK Foresight Committee, etc. He chairs the Index Advisory Group of Euronext, is a member of the Scientific Committee of the French regulator (AMF), and has been part of the Consultative Workgroup on Financial Innovation of the European Authority (ESMA). Moreover, Charles-Albert received the 2016 Best Paper Award in Finance from Europlace Institute for Finance (EIF) and published more than fifty academic papers and book chapters. He co-authored the book “Market Microstructure in Practice” (World Scientific Publisher, 2nd ed 2018), analyzing the main features of modern markets. With a PhD in machine learning, Charles-Albert is chairing the “Finance and Insurance Reloaded” transverse research program of the Louis Bachelier Institute, this program explores the influence of new technologies (from blockchain to artificial intelligence) on our industries.
Machine Learning for Financial Markets : three representative examples
As a generic technology, machine learning has numerous secondary innovations in a lot of sectors. In this talk I will discuss the innovations emerging (or to appear) for financial markets. I will provide one example of online learning on trading flows (for real-time Dark Pools selection), another one of now casting (satellite images processing for crop quality prediction), and the last one will be on human-machine interface (decision support to monitor hundreds of algorithms). This last example is significant of one of the greatest challenge linked to the use of AI: when and how to give back the control to humans, so that then can really take an enlighten decision.