Eigen Memory Trees
This work addresses memory efficiency in sequential learning for AI systems, offering a novel hybrid approach with strong performance gains.
The paper tackles the problem of online memory for sequential learning by introducing Eigen Memory Trees (EMT), which store data in a binary tree and use principal components for routing, achieving efficient logarithmic access and outperforming existing methods, with validation on 206 datasets from OpenML.
This work introduces the Eigen Memory Tree (EMT), a novel online memory model for sequential learning scenarios. EMTs store data at the leaves of a binary tree and route new samples through the structure using the principal components of previous experiences, facilitating efficient (logarithmic) access to relevant memories. We demonstrate that EMT outperforms existing online memory approaches, and provide a hybridized EMT-parametric algorithm that enjoys drastically improved performance over purely parametric methods with nearly no downsides. Our findings are validated using 206 datasets from the OpenML repository in both bounded and infinite memory budget situations.