Cheng-Yuan Liou

Autoencoder for words

This paper presents a training method that encodes each word into a different vector in semantic space and its relation to low entropy coding. Elman network is employed in the method to process word sequences from literary works. The trained codes possess reduced entropy and are used in ranking, indexing, and categorizing literary works. A modification of the method to train the multi-vector for each polysemous word is also presented where each vector represents a different meaning of its word. These multiple vectors can accommodate several different meanings of their word. This method is applied to the stylish analyses of two Chinese novels, Dream of the Red Chamber and Romance of the Three Kingdoms.

Modeling word perception using the Elman network

This paper presents an automatic acquisition process to acquire the semantic meaning for the words. This process obtains the representation vectors for stemmed words by iteratively improving the vectors, using a trained Elman network. Experiments performed on a corpus composed of Shakespeares writings show its linguistic analysis and categorization abilities.

Manifold Construction by Local Neighborhood Preservation

This work presents a neighborhood preservation method to construct the latent manifold. This manifold preserves the relative Euclidean distances among neighboring data points. Its computation cost is close to the linear algorithm and its performance in preserving the local relationships is promising when we compared it with the methods, LLE and Isomap.

Handprinted character recognition based on spatial topology distance measurement

In this work we present a self-organization matching approach to accomplish the recognition of handprinted characters drawn with thick strokes. This approach is used to flex the unknown handprinted character toward matching its object characters gradually. The extracted character features used in the self-organization matching are center loci, orientation, and major axes of ellipses which fit the inked area of the patterns. Simulations provide encouraging results using the proposed method.

Resolving Hidden Representations

This paper presents a novel technique to separate the pattern representation in each hidden layer to facilitate many classification tasks. This technique requires that all patterns in the same class will have near representions and the patterns in different classes will have distant representions. This requirement is applied to any two data patterns to train a selected hidden layer of the MLP or the RNN. The MLP can be trained layer by layer feedforwardly to accomplish resolved representations. The trained MLP can serve as a kind of kernel functions for categorizing multiple classes.

Dynamic Positron Emission Tomography Data-Driven Analysis Using Sparse Bayesian Learning

A method is presented for the analysis of dynamic positron emission tomography (PET) data using sparse Bayesian learning. Parameters are estimated in a compartmental framework using an over-complete exponential basis set and sparse Bayesian learning. The technique is applicable to analyses requiring either a plasma or reference tissue input function and produces estimates of the systems macro-parameters and model order. In addition, the Bayesian approach returns the posterior distribution which allows for some characterisation of the error component. The method is applied to the estimation of parametric images of neuroreceptor radioligand studies.

Conformal self-organization for continuity on a feature map

The self-organization model with a conformal-mapping adaptation is studied in this work. This model is designed to provide conformal transformation to meet the conformality requirement in biological morphology and geometrical surface mapping. This model spans the network field in the input space where topological conformality is preserved. The converged network provides not only the organized clustering features of the input but also a specific mapping representation. This facilitates the Kohonens self-organization model in exploring the input in a continuous conformality sense. Simulations for morphing applications are described.

Error tolerant associative memory

Abstract. We present a new approach to enlarging the basin of attraction of associative memory, including auto-associative memory and temporal associative memory. The memory trained by means of this method can tolerate and recover from seriously noisy patterns. Simulations show that this approach will greatly reduce the number of limit cycles.

Conformal self-organizing map for a genus-zero manifold

This paper presents the implementation of a surface mesh on a genus-zero manifold with 3D scattered data of sculpture surfaces using the conformal self-organizing map (CSM). It starts with a regular mesh on a sphere and gradually shapes the regular mesh to match its object’s surface by using the CSM. It can drape a uniform mesh on an object with a high degree of conformality. It accomplishes the surface reconstruction and also defines a conformal mapping from a sphere to the object’s manifold.

Self-organization using Potts models

In this work, we use Potts neurons for the competitive mechanism in a self-organization model. We obtain new algorithms on the basis of a Potts neural network for coherent mapping, and we remodel the Durbin algorithm and the Kohonen algorithm with mean field annealing. The resulting dimension-reducing mappings possess a highly reliable topology preservation such that the nearby elements in the parameter space are ordered as similarly as possible on the cortex-like map, and the objective function costs between neighboring cortical points are as smooth as possible. The proposed Potts neural network contains two sets of interactive dynamics for two kinds of mappings, one from the parameter space to the cortical space and the other in the reverse way. We present a theoretical approach to developing self-organizing algorithms with a novel decision principle for competitive learning. We find that one Potts neuron is able to implement the Kohonen algorithm. Both implementation and simulation results are encouraging.