Intelligence, please.
Applied AI research grounded in formal methods.
Multiagent systems. Belief revision. Knowledge representation.
Scroll
Intelligence, please.
Applied AI research grounded in formal methods.
Multiagent systems. Belief revision. Knowledge representation.
Architectures for autonomous agents that coordinate, negotiate, and reason. Protocol design, communication semantics, and formal interaction models — not prompt chains calling prompt chains.
Ontological engineering with formal semantics. Description logics, graph-based knowledge structures, and constraint validation that machines can reason over — not just retrieve from.
How agents rationally update what they know. Theory change under formal guarantees, epistemic entrenchment, and belief propagation across distributed knowledge bases.
Every framework, tool, and dataset we build ships publicly. Reproducible research, documented APIs, and community-driven development. If you can't reproduce it, it's not science.
Most AI systems are built on intuition and iteration. We start from formal foundations — the same principles that let distributed systems reach consensus and databases maintain consistency.
Belief revision gives agents real epistemology. Ontologies give knowledge real structure. The result is AI that can explain what it knows and why it changed its mind.