Abstract: This talk presents results on two aspects of risk-aware reinforcement learning. The first part looks at how sequences of cash-flows may be optimized using dynamically time-consistent convex risk-measures in finite time horizon. We do this, in a model-free manner, using neural net parameterized policies and develop an actor-critic approach to solving the resulting dynamic programming equations with recursive gradients. In the second part of the talk, to allow agents to express a wide variety of risk-reward profiles, we assess the value of a policy using rank dependent expected utility (RDEU). RDEU allows the agent to seek gains, while simultaneously protecting themselves against downside events. To robustify optimal policies against model uncertainty, we assess a policy not by its distribution, but rather, by the worst possible distribution that lies within a Wasserstein ball around it. In both parts, we illustrate the efficacy of the approaches on various financial problems.

Bio: Professor Jaimungal is the current Director of the professional Masters of Financial Insurance program in the Department of Statistical Sciences, and he teaches in the Mathematical Finance Program at the University of Toronto, as well as the PhD and MSc programs in the Department of Statistical Sciences.

Professor Jaimungal is the former Chair for the SIAM activity group in Financial Mathematics and Engineering (SIAG/FM&E), and a Managing Editor of Quantitative Finance, an Associate Editor for the SIAM Journal on Financial Mathematics (SIFIN), Frontiers of Mathematical Finance, the International Journal of Theoretical and Applied Finance (IJTAF), Journal of Dynamics and Games, Journal of Risks, and Argo. As well, He was a founding board member of the Commodities and Energy Markets Association and now serve on its advisory board.

Professor Jaimungal is a Fields-CQAM lab leader for the Systemic Risk Analytics lab. The lab’s purpose is to look at both data-driven and model-driven research surrounding issues of systemic risk in its various guises ranging from micro-structure to inter-bank networks. His current focus is on the study of intra-day volatility and its effects in FX and related markets using statistical and machine learning methodologies.

Professor Jaimungal has been named as a Fellow at Fields Institute for Mathematical Sciences since 2020. He is also a SIGEST award winner awarded by SIAM for the best paper published during 2013-2017 in SIAM J. Finan. Math.

Meeting Recording: https://ucsb.zoom.us/rec/share/CuD1iWZn_L7V98uYTig4Ol0y-DxwqerhmitCeC2j_GQrsfJcbUeUL8oIEKis1MN6.0hLXj1XWE8Vd0Zfr

Access Passcode: Ulrq+T5h