The purpose of this paper is to propose and develop a new conceptual framework for
approximate Dynamic Programming (DP) and Reinforcement Learning (RL). This framework …

In this paper we aim to provide analysis and insights (often based on visualization), which
explain the beneficial effects of on-line decision making on top of off-line training. In …

The purpose of this paper is to present a new numerical approach for finding the early
exercise boundary of discounted American options, which payment structure is generalized …

The aim of this study is to explore the behavior of perpetual game put options, also known as
cancellable puts. Their main characteristic is the opportunity of the buyer and the seller to …

In this thesis, the numerical solution of three different classes of problems have been
studied. Specifically, new techniques have been proposed and their theoretical analysis has …

We report on a general purpose method for the scalar Stefan problem inspired by the
standard boundary updating method used in several existence proofs. By suitably modifying …

We investigate the asymptotic behaviors of American maximum options with dividendpaying
assets near maturity. Using the exercise conditions of American options, we obtain the …

Derivatives are one of the most important instruments against the financial risks. They are
based on some underlying object which can be a single asset, portfolio of assets, financial …

The purpose of this paper is to propose and develop a new conceptual framework for
approximate Dynamic Programming (DP) and Reinforcement Learning (RL). This framework …