Offline Reinforcement Learning (RL) aims to learn a near-optimal policy from a fixed dataset
of transitions collected by another policy. This problem has attracted a lot of attention …

We study the performance of stochastic first-order methods for finding saddle points of
convex-concave functions. A notorious challenge faced by such methods is that the gradients can …

We propose a new stochastic primal-dual optimization algorithm for planning in a large
discounted Markov decision process with a generative model and linear function approximation. …

… First, to my Grandmother, Catherine Okolo; thank you so much for your prayers and presence.
You … And finally, to my Grandfather, James Okolo. Thank you for your proud, protective and …

We study offline Reinforcement Learning in large infinite-horizon discounted Markov Decision
Processes (MDPs) when the reward and transition models are linearly realizable under a …

Real-world sequential decision-making tasks are often characterized by large state spaces
and unknown dynamics. In this setting, an efficient method is tasked to learn a near-optimal …

Covid-19 pandemic is a very harsh experience to the world at large and Africa in particular.
The strategies to prevent it are at the same time uncomfortable for all and sundry. This harsh …

We study a linear programming (LP) approach to planning in large Markov Decision
Processes (MDPs), addressing some well-known tractability issues of traditional LP-based …

Climate change has emerged as a critical driver of internal migration, particularly in Northern
Nigeria where environmental stressors such as drought, desertification, and flooding have …

The present state of the economy after COVID-19 with eight months lockdown brought about
no production in the country; whereby people were feeding on their savings and capital, …