Hussein Fellahi

Predictive and prescriptive analytics to bridge the gap between segmentation and prediction for real-world applications Hussein Fellahi · Follow Published in Towards Data Science · 11 min read · 7 hours ago — Photo by NASA Hubble Space Telescope on Unsplash Introduction: In many real-world machine learning tasks, the population being studied is often diverse and heterogeneous. This variability presents unique challenges, particularly in regression and classification tasks where a single, generalized model may fail

A novel tractable and interpretable algorithm to learn from expert demonstrations Hussein Fellahi · Follow Published in Towards Data Science · 12 min read · 11 hours ago — Photo by Andrea De Santis on Unsplash Introduction Imitation Learning has recently gained increasing attention in the Machine Learning community, as it enables the transfer of expert knowledge to autonomous agents through observed behaviors. A first category of algorithm is Behavioral Cloning (BC), which aims to

Theoretical derivation of the Robust Counterpart of Markov Decision Processes (MDPs) as a Linear Program (LP) Hussein Fellahi · Follow Published in Towards Data Science · 8 min read · 10 hours ago — Photo by ZHENYU LUO on Unsplash Introduction Markov Decision Processes are foundational to sequential decision-making problems and serve as the building block for reinforcement learning. They model the dynamic interaction between an agent having to make a series of actions and