Dynamics of Projective Adaptive Resonance Theory Model: The Foundation of PART Algorithm

Cao, Yongqiang; Wu, Jianhong

doi:10.1109/tnn.2004.824261

Cited by 49 publications

(38 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1a) and PART (Fig. 1b) networks are governed by the equations that describe the short term (STM) and long term (LTM) memory traces together with a similarity condition [25], [26]. STM corresponds to the type of memory that can be readily reset without leaving an enduring trace.…”

Section: Art Neural Networkmentioning

confidence: 99%

“…Following [25], [26], the STM equations for layers F 1 and F 2 in a PART neural network (Fig. 1b) are…”

Section: A Short-term Memory (Stm) Equationsmentioning

confidence: 99%

“…Following the convention of [22], we use the index i to refer to the F 1 nodes and j to refer to the F 2 nodes. We take A = 1, B = 0 and C > 0 [25], [26]. The terms J + and J − give the F 2 layer self-excitation and self-inhibition, and have their typical form…”

Section: A Short-term Memory (Stm) Equationsmentioning

confidence: 99%

“…However, as the number of dimensions becomes large it becomes increasingly unlikely that data is similar on all dimensions. Consequently ART neural network perfom poorly on realistic large data sets [24] - [26]. This problem cannot be solved by performing a dimension reduction before analysis using techniques such as Principal Components Analysis (PCA) [29].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Clustering neural spike trains with transient responses

Hunter¹,

Wu²,

Milton

2008

2008 47th IEEE Conference on Decision and Control

View full text Add to dashboard Cite

Abstract-The detection of transient responses, i.e. nonstationarities, that arise in a varying and small fraction of the total number of neural spike trains recorded from chronically implanted multielectrode grids becomes increasingly difficult as the number of electrodes grows. This paper presents a novel application of an unsupervised neural network for clustering neural spike trains with transient responses. This network is constructed by incorporating projective clustering into an adaptive resonance type neural network (ART) architecture resulting in a PART neural network. Since comparisons are made between inputs and learned patterns using only a subset of the total number of available dimensions, PART neural networks are ideally suited to the detection of transients. We show that PART neural networks are an effective tool for clustering neural spike trains that is easily implemented, computationally inexpensive, and well suited for detecting neural responses to dynamic environmental stimuli.

show abstract

Section: Art Neural Networkmentioning

confidence: 99%

“…Following [25], [26], the STM equations for layers F 1 and F 2 in a PART neural network (Fig. 1b) are…”

Section: A Short-term Memory (Stm) Equationsmentioning

confidence: 99%

Section: A Short-term Memory (Stm) Equationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Clustering neural spike trains with transient responses

Hunter¹,

Wu²,

Milton

2008

2008 47th IEEE Conference on Decision and Control

View full text Add to dashboard Cite

show abstract

“…Recently, Cao and Wu (2004) have developed a very effective high-dimensional network called Projective ART (PART), based on the assumption that the model equations of PART, a large scale and singularly perturbed system of differential equations coupled with a reset mechanism, have quite regular computational performance.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Resonance Theory Training Parameters: Pretty Good Sets

Eldos¹

2010

Journal of Computer Science

View full text Add to dashboard Cite

Problem statement: ART1 artificial neural networks offer good tools for test clustering, where no expert is needed if the system is well trained. However, having no output reference for the input patterns makes it hard to judge the quality of the training. Moreover, the performance depends to a great extent on a set of training parameters. Designers follow some recommendations or depend on their expertise in finding good sets with no performance guarantees. Many methods were proposed; from greedy methods offering quick and acceptable solutions to evolutionary algorithms offering suboptimal sets of parameters. While the evolutionary algorithms are a good choice for quality, the computational cost is large even for an offline process; after all, computing resources are not for free. Approach: We introduced a method for selecting a set of parameters that yields a comparable performance and robust operation, with relatively low cost compared to the evolutionary methods. This method located a suitable set through repetitive portioning of the range, by considering the best subset for the next iteration. Results: Tests have shown that performance comparable with the computationally intensive evolutionary methods could be achieved in much less time. Conclusion: The repetitive portioning method for finding a good set of training parameters is very cost effective and yields good performance.

show abstract

Pattern Recognition

2010

Image Processing and Pattern Recognition

View full text Add to dashboard Cite

Dynamics of Projective Adaptive Resonance Theory Model: The Foundation of PART Algorithm

Cited by 49 publications

References 27 publications

Clustering neural spike trains with transient responses

Clustering neural spike trains with transient responses

Adaptive Resonance Theory Training Parameters: Pretty Good Sets

Pattern Recognition

Contact Info

Product

Resources

About