An evaluation of the neocognitron
نویسندگان
چکیده
We describe a sequence of experiments investigating the strengths and limitations of Fukushima's neocognitron as a handwritten digit classifier. Using the results of these experiments as a foundation, we propose and evaluate improvements to Fukushima's original network in an effort to obtain higher recognition performance. The neocognitron performance is shown to be strongly dependent on the choice of selectivity parameters and we present two methods to adjust these variables. Performance of the network under the more effective of the two new selectivity adjustment techniques suggests that the network fails to exploit the features that distinguish different classes of input data. To avoid this shortcoming, the network's final layer cells were replaced by a nonlinear classifier (a multilayer perceptron) to create a hybrid architecture. Tests of Fukushima's original system and the novel systems proposed in this paper suggest that it may be difficult for the neocognitron to achieve the performance of existing digit classifiers due to its reliance upon the supervisor's choice of selectivity parameters and training data.
منابع مشابه
Evaluation of Two Neocognitron-type Models for Recognition of Rotated Patterns
We examine the number of cells and execution time taken to correctly recognize rotated patterns in two models: a rotation-invariant neocognitron (RNeocognitron) and a neocognitron-type model (TDR-Neocognitron) which recognizes rotated patterns by use of an associative recalled pattern. In numerical simulations handwritten patterns in CEDER database are used for training and evaluation of recogn...
متن کاملAn alternative approach for the evaluation of the neocognitron
The neocognitron network is analysed from the point of view of the contribution of the di erent layers to the nal classi cation. A variation to the neocognitron which gives improved performance is suggested. This variant combines the low level feature extraction capabilities of the initial layers with alternative classi ers such as LVQ and Class Based Means Clustering. This is shown to give per...
متن کاملSimulation of the Neocognitron on a CCD Parallel Processing Architecture
The neocognitron is a neural network for pattern recognition and feature extraction. An analog CCD parallel processing architecture developed at Lincoln Laboratory is particularly well suited to the computational requirements of shared-weight networks such as the neocognitron, and implementation of the neocognitron using the CCD architecture was simulated. A modification to the neocognitron tra...
متن کاملNeocognitron's Parameter Tuning by Genetic Algorithms
The further study on the sensitivity analysis of Neocognitron is discussed in this paper. Fukushima's Neocognitron is capable of recognizing distorted patterns as well as tolerating positional shift. Supervised learning of the Neocognitron is fulfilled by training patterns layer by layer. However, many parameters, such as selectivity and receptive fields are set manually. Furthermore, in Fukush...
متن کاملIs the Neocognitron Capable of State-of-the-art Digit Recognition?
We describe a series of experiments that evaluate the performance of Fukushima's neocognitron using a database of handwritten ZIP code digits. A number of improvements to the original neocognitron were proposed and implemented, resulting in a peak performance of 85.54% correct classiication, with 95.69% reliability. This result suggests that, with appropriate modiications, the neocognitron is a...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- IEEE transactions on neural networks
دوره 8 5 شماره
صفحات -
تاریخ انتشار 1997