Fixed-weight on-line learning.

Conventional artificial neural networks perform functional mappings from their input space to their output space. The synaptic weights encode information about the mapping in a manner analogous to long-term memory in biological systems. This paper presents a method of designing neural networks where recurrent signal loops store this knowledge in a manner analogous to short-term memory. The synaptic weights of these networks encode a learning algorithm. This gives these networks the ability to dynamically learn any functional mapping from a (possibly very large) set, without changing any synaptic weights. These networks are adaptive dynamic systems. Learning is online continually taking place as part of the network's overall behavior instead of a separate, externally driven process. We present four higher order fixed-weight learning networks. Two of these networks have standard backpropagation embedded in their synaptic weights. The other two utilize a more efficient gradient-descent-based learning rule. This new learning scheme was discovered by examining variations in fixed-weight topology. We present empirical tests showing that all these networks were able to successfully learn functions from both discrete (Boolean) and continuous function sets. Largely, the networks were robust with respect to perturbations in the synaptic weights. The exception was the recurrent connections used to store information. These required a tight tolerance of 0.5%. We found that the cost of these networks scaled approximately in proportion to the total number of synapses. We consider evolving fixed weight networks tailored to a specific problem class by analyzing the meta-learning cost surface of the networks presented.

Sizing dielectric spheres and cylinders by aligning measured and computed resonance locations: algorithm for multiple orders.

An algorithm for determining the size of dielectric spheres and cylinders by aligning measured and computed resonance locations is presented. The orders of the resonance locations need not be known a priori. The algorithm is applicable to several types of scattering and emission spectra of spheres and cylinders if the index of refraction including dispersion is known and uniform, or nearly uniform, throughout the sphere or cylinder. The algorithm performs well when tested with groups of computed resonance locations of spheres (synthetic data) and with measured fluorescence emission spectra of spheres exhibiting as many as 5 orders of resonance.