Kazuki Irie
Assistant Professor of Computer Science
Department of Computer Science &
Wu Tsai Institute
Yale University
WTI Office: Room 1122, 100 College St, New Haven, CT 06510
Email:
kazuki.irie@yale.edu
I am hiring at all levels — Ph.D. students, postdocs, research assistants, and interns.
Google Scholar / GitHub / dblp / OpenReview / arXiv / Twitter
Short Bio: Kazuki Irie is a tenure-track Assistant Professor of Computer Science and a Wu Tsai Investigator at Yale University. His COSMIC Lab conducts fundamental research in deep learning, develops cognitive machinery and memory systems, and explores novel computational hypotheses in cognitive neuroscience and beyond. Before joining Yale, he was a postdoctoral researcher in the Department of Psychology at Harvard University and at the “Swiss AI Lab IDSIA” at the University of Lugano, Switzerland, where he also served as a lecturer teaching deep learning. He received his PhD in Computer Science from RWTH Aachen University in Germany, after completing undergraduate and Master’s studies in Applied Mathematics at École Centrale Paris and ENS Cachan in France.
Research
I conduct fundamental research in deep learning aimed at developing the foundational building blocks for ever more capable artificial intelligence (AI) systems. In particular, I work on general-purpose sequence-processing neural networks and learning algorithms. As such, my projects span a variety of domains—including supervised/unsupervised learning in language, vision, and algorithmic tasks, as well as reinforcement learning in partially observable environments such as video games—without committing to a single application area (although my original training was in language modeling). My focus is on achieving sophisticated cognitive abilities such as continual learning, (meta-)metalearning, few-shot learning, analogical and compositional generalization, and efficient memory-guided exploration—through a unified approach of developing advanced memory algorithms that better leverage past experiences to improve future behavior. I call this “memory-centric deep learning”.
I am also putting forward the idea that advances in AI algorithms can offer computational models that can serve as novel hypotheses in the study of natural intelligence—hypotheses optimized by computer scientists towards their own objectives of efficiency and problem solving that may yield unconventional insights that differ from traditional approaches in cognitive science and neuroscience. Conversely, thinking about human intelligence provides insights into elements currently missing from state-of-the-art AI systems, while also highlighting the unique strengths of machine intelligence—reminding me the reasons why I like both humans and machines. It is also this comparative and multidisciplinary perspective on intelligence that guides my vision of life-centric AGI—artificial general intelligence (AGI) that supports humans in addressing open challenges and enhances the well-being of people, animals, and the broader living world.
Prospective students
Multiple open positions at all levels will be announced soon: Ph.D. students (application deadline in Dec. 2026), postdocs, research assistants, and interns. More details coming soon.