Development of low entropy coding in a recurrent network.

In this paper we present an unsupervised neural network which exhibits competition between units via inhibitory feedback. The operation is such as to minimize reconstruction error, both for individual patterns, and over the entire training set. A key difference from networks which perform principal components analysis, or one of its variants, is the ability to converge to non-orthogonal weight values. We discuss the network's operation in relation to the twin goals of maximizing information transfer and minimizing code entropy, and show how the assignment of prior probabilities to network outputs can help to reduce entropy. We present results from two binary coding problems, and from experiments with image coding.

The length-weight relationship in Sardinella sirm Wal from the Andaman Sea

Preliminary details are given of studies conducted regarding the length-weight relationship of Sardinella sirm from the Andaman Sea, a species of economic importance in the area. Results show that this species grows in proportion to length, weight and girth; the group of fish studied was composed of the same population and did not reveal any subspecies of sub-variety.

Length-weight relationship of Upeneus sulphureus (Cuvier) collected off Maharashtra coast

The length-weight relationship of Upeneus sulphureus was studied on the basis of 125 females and 103 males, collected during the 27th cruise of research vessel M.V. Saraswati in the month of September, 1984. The correlation coefficient was found significant at 5% and 1% level respectively for male and female. In both the sexes, the regression co-efficient was found higher than three. Analysis of covariance was also carried out for the F test and F distribution.