Research
Open-source work across information geometry, emergent computation,
spatiotemporal signal processing, and acoustic localization. All four
projects deal with the same general problem: recovering coherent
structure from incomplete or noisy data.
Everything here is open-source. If any of it overlaps with your own
work, reach out.
Evidence Holonomy
Information geometry Python Library
Python library implementing KL-rate holonomy estimators, entropy production estimation, and arrow-of-time detection. Reformulates existing information-geometric ideas about irreversibility into a usable tool with CLI and demos across audio, images, video, sensor data, and financial time series.
If you work with sequential data and want to measure directional asymmetry (does this process look different forwards vs. backwards?), this gives you a geometric way to quantify that. The arrow-of-time demos show it working across very different data types.
View on GitHub ↗
Embodied Field Intelligence
Emergent computation Python · Jupyter Library
Uses cellular automata as a substrate for real-time distributed intelligence in embodied agents. Instead of centralized planning, agents navigate and learn through local field operations: diffusion-based scent trails for navigation, visit trails and novelty detection for memory, and local Oja/BCM/slowness-based learning for schema acquisition. All computation stays CA-native, meaning every behavior emerges from local cell updates over cognitive fields.
A way to get intelligent-looking behavior (exploration, avoidance, adaptation) without anything that resembles a brain. Relevant if you're interested in minimal architectures for embodied agents, or in how far local field dynamics can go before you need centralized reasoning.
View on GitHub ↗
CoCore
Signal reconstruction Python Private
Exploring graph-based approaches to reconstructing coherent spatial fields from sparse, noisy sensor data. The graph structure handles three related problems: filtering measurement noise, identifying which sensors are drifting or broken, and interpolating between sensors to recover the full spatial picture.
Potentially applicable to environmental monitoring, infrastructure health assessment, or any domain with irregular sensor placement. The graph formulation means the approach doesn't require a uniform grid, which is where most classical interpolation methods start.
Private repository
Acoustic Event Locator
TDOA multilateration Python Library · working prototype
Library for locating impulsive acoustic events (gunshots, explosions, impacts) using time-difference-of-arrival from unsynchronized consumer video. You feed it video files from different phones with different start times and positions. It extracts the impulsive event from each audio track, estimates relative time offsets, and multilaterates the source location.
Built for forensic and public safety applications. After an incident, bystander video often exists but nobody knows where exactly the event originated. This library can answer that from the audio alone, without purpose-built sensor infrastructure.
View on GitHub ↗