Alexey Skrynnik

Papers

Total Citations

H-Index

About

Alexey Skrynnik is a researcher at the forefront of artificial intelligence, specializing in hierarchical reinforcement learning, multi-agent systems, and pathfinding under uncertainty. His work bridges the gap between classical planning and modern learning-based approaches, particularly in complex, stochastic environments. Skrynnik’s most-cited paper, “Forgetful experience replay in hierarchical reinforcement learning from expert demonstrations” (2021, 22 citations), introduces a novel method that improves sample efficiency by selectively forgetting irrelevant experiences, enabling more robust learning from limited expert data. In “Pathfinding in stochastic environments: learning vs planning” (2022, 10 citations), he tackles the challenge of navigating mobile robots through environments where obstacles can appear and disappear randomly, proposing a stochastic formulation that outperforms deterministic planners. His work “Planning and Learning in Multi-Agent Path Finding” (2022, 9 citations) addresses the scalability issues of coordinating numerous robots—such as those in automated warehouses—by combining learning and planning to achieve efficient, collision-free navigation. Skrynnik’s contributions are vital for advancing autonomous systems in dynamic, real-world settings, and his research continues to influence both theoretical and applied AI.