I solve problems in everything from computer vision to deep reinforcement learning(RL) applied to algorithmic trading. Strong background in linear algebra, probability, set theory, and other mathematics. Utilizing lie groups in the context of theoretical physics in my free time(gauge theory). Mainly work in python/tensorflow.
My knowledge of ML algorithms spans from classic methods(linear CoF, random forests, SVMs), to cutting-edge techniques(capsule networks, ANNs learning to knowledge graphs for probabilistic reasoning, pointer networks). Ergo, I can reason about the optimality of a particular architecture in a given problem domain.
While I consider myself primarily focused on reinforcement learning and top-down AGI, I have implemented multiple recommendation systems successfully, developed novel computer vision and NLP solutions, as well as tabular problems. Additionally, many solutions involved “big data” solutions via Spark, Hadoop, and others.
Further, I am developing state-of-the-art frameworks and architectures in the pursuit of Strong/General AI. Progress in this area has led me to advancements such as combining deep reinforcement learning algorithms with a version of OpenCog’s Bayesian Meta-Optimizing Semantic Evolutionary Search and atomspace to craft the beginnings of generalized learning framework and formalize it mathematically.