systems and reinforcement
Reforcemind explores the fundamental post-training bottlenecks of Vision-Language-Action models.
We look for people who reason from first principles and operate with high agency. If you care deeply about the systems you build and the way you build them, we would love to hear from you.
Multi-Agent Reinforcement Learning (MARL) cybersecurity simulator
Python 2
A hpc LLVM Pass extracting semantic Control-Data Flow Graphs (CDFG) from Intermediate Representation for Graph Neural Networks. Enables cross-language code retrieval and clone detection beyond toke…
Python 1
A framework for end-to-end AI inference optimization: from model parsing and graph IR, through graph and kernel optimizations, to hardware profiling, auto-tuning, and visualization.
Python
Event-Driven Continuous-Time Graph MARL for Asynchronous Cyber Defense in NetForge_RL. A framework for training autonomous cyber defenders in a continuous-time POSMDP environment. Features Neural ODE temporal dynamics, GAT spatial reasoning, and a Sim2Real evaluation bridge against live CVE payloads.
Implementation of Epistemic Time-Dilation MAPPO (ETD-MAPPO). A compute-aware MARL framework where agents autonomously modulate their execution frequency based on uncertainty to reduce inference overhead.
A hpc LLVM Pass extracting semantic Control-Data Flow Graphs (CDFG) from Intermediate Representation for Graph Neural Networks. Enables cross-language code retrieval and clone detection beyond token-based approaches.
A framework for end-to-end AI inference optimization: from model parsing and graph IR, through graph and kernel optimizations, to hardware profiling, auto-tuning, and visualization.
This organization has no public members. You must be a member to see who’s a part of this organization.
Loading…
Loading…
