13 projects
bloomer
Bloomer — brain-mimetic memory for AI. Seven biological principles, one Python library.
meta-learning-toolkit
Production-ready meta-learning algorithms with research-accurate implementations of MAML, Prototypical Networks, and test-time compute scaling
qualitative-reasoning
Qualitative physics simulation and causal reasoning systems
universal-learning
Solomonoff induction and AIXI for universal artificial intelligence
tensor-product-binding
Structured knowledge representation in neural networks via tensor products
holographic-memory
Compositional memory with fixed-size vectors via circular convolution (Holographic Reduced Representations)
information-bottleneck
Optimal compression-prediction tradeoffs for principled feature extraction
sparse-coding
Research-faithful sparse coding with Olshausen-Field paper-exact mode and modern L1/FISTA.
inductive-logic-programming
FOIL and Progol algorithms for learning logical rules from examples
self-organizing-maps
Kohonen Self-Organizing Maps for unsupervised learning and visualization
reservoir-computing-benedictchen
Echo State Networks and Liquid State Machines for temporal pattern recognition
meta-learning-research
Advanced meta-learning algorithms including test-time compute scaling, MAML variants, and few-shot learning
m3u8-achiver
Download m3u8 files as-is
