K. Cardenas

About

K. Cardenas is a researcher in artificial intelligence, specializing in deep reinforcement learning and sequential decision-making. Their major contribution lies in developing efficient deep Q-learning strategies that enable agents to learn intelligent policies directly from high-dimensional sensory inputs, such as video feeds. In their most-cited work, "An Efficient Deep Q-learning Strategy for Sequential Decision-making in Game-playing" (2022, 5 citations), Cardenas demonstrated a novel approach for training an AI to play tic-tac-toe by using a convolutional neural network to produce stable, sparse representations of game states. This work highlights their focus on bridging the gap between raw perceptual data and effective action selection, a key challenge in modern AI. Although early in their career, Cardenas’s research has implications for robotics, autonomous systems, and interactive AI, where learning from visual inputs is critical. Their work is a valuable resource for students and researchers exploring efficient reinforcement learning techniques for real-world applications.

Research Focus

Key Achievements

Top Papers

1. 1
   An Efficient Deep $Q$-learning Strategy for Sequential Decision-making in Game-playing

   5 citations · 2022

Key Collaborators

Contact & Links