Holographic Neural Architectures
This work addresses representation learning for AI, with potential applications in biological domains where data is scarce or noisy, though it appears incremental as it builds on existing representation learning concepts.
The paper tackles representation learning by introducing Holographic Neural Architectures (HNAs), which derive holographic representations from training sets and enable exploration along a continuous dimension, resulting in generative networks, state-of-the-art regression models, and high noise resistance.
Representation learning is at the heart of what makes deep learning effective. In this work, we introduce a new framework for representation learning that we call "Holographic Neural Architectures" (HNAs). In the same way that an observer can experience the 3D structure of a holographed object by looking at its hologram from several angles, HNAs derive Holographic Representations from the training set. These representations can then be explored by moving along a continuous bounded single dimension. We show that HNAs can be used to make generative networks, state-of-the-art regression models and that they are inherently highly resistant to noise. Finally, we argue that because of their denoising abilities and their capacity to generalize well from very few examples, models based upon HNAs are particularly well suited for biological applications where training examples are rare or noisy.