One-to-one Mapping between Stimulus and Neural State: Memory and Classification
This addresses memory and classification mechanisms in neural networks, but appears incremental as it builds on existing synaptic plasticity theories.
The paper tackles the problem of how neural networks can memorize external stimuli through synaptic plasticity, proposing that under certain constraints, a one-to-one mapping exists between stimulus and synaptic strength at a fixed point, and introduces a biological classifier based on this mapping.
Synaptic strength can be seen as probability to propagate impulse, and according to synaptic plasticity, function could exist from propagation activity to synaptic strength. If the function satisfies constraints such as continuity and monotonicity, neural network under external stimulus will always go to fixed point, and there could be one-to-one mapping between external stimulus and synaptic strength at fixed point. In other words, neural network "memorizes" external stimulus in its synapses. A biological classifier is proposed to utilize this mapping.