Who
We are an AI research lab focused on algorithmic and architectural breakthroughs to predict real-world change.
Why
The current machine learning paradigm struggles to understand change in non-stationary environments like financial markets. These systems require a continuous, real-time understanding of change, rather than a reliance on historical patterns which creates a critical detection lag in time-critical systems.
How
A proprietary zero-context, continual learning architecture that autonomously constructs high-dimensional world models to identify signal within noise.
By operating in non-stationary data streams with a continuous understanding of change, our architecture detects causal directional shifts before traditional statistical models reach significance. This is achieved with zero reliance on training data, pre-defined context or periodic retraining.
What
We provide a lead-time advantage by identifying the causal directional shift in non-stationary environments before they are detected in traditional statistical metrics.
Team
Based in London. Our team, investors, and advisory board bring expertise from Harvard University, The House of Commons, Lloyds Bank, JLL, Oppenheimer Group, Incu Global and Zapier.
Latest
Demonstrating our algorithmic approach for non-stationary environments at the Society for Industrial and Applied Mathematics.
Modelling global markets as a single causal information network to mitigate black swan events.
A predictive risk framework for G-SIBs that identifies tail-risk emergence before it manifests in traditional Expected Shortfall metrics.
Supported By
