This paper presents a method for effectively using unlabeled sequential data in the learning of hidden Markov models (HMMs). With the conventional approach, class labels for unlabeled data are assigned deterministically by HMMs learned from labeled data. Such labeling often becomes unreliable when the number of labeled data is small. We propose an extended Baum-Welch (EBW) algorithm in which th...

An experiment was conducted to study energy retained (TER) as fat (TERF) and protein (TERP) in 3 strains of birds with different growth rate; commercial broilers, Barred Plymouth Rock, and Leghorns. Birds were fed ad libitum a diet providing 3,100 kcal of AMEn/kg and 20% CP from 0 to 42 d. Body composition, TER, TERF, and TERP were determined at 0, 7, 10, 15, 19, 23, 28, 33, 37, and 42 d of age...

The aim of study was to determine the energy requirements for maintenance and growth of forty-one Saanen, intact male kids with initial body weight (BW) of 5.12±0.19 kg. The baseline (BL) group consisted of eight kids averaging 5.46±0.18 kg BW. An intermediate group consisted of six kids, fed for ad libitum intake, that were slaughtered when they reached an average BW of 12.9±0.29 kg. The remai...

OBJECTIVE To determine the prevalence of excess body weight and abdominal obesity and assess the association with socioeconomic, demographic and behavioral factors among adults of Maranhão, Brazil. METHODS Multistage cluster sample investigated through a household survey on 1,005 adults aged 20 to 59 years (393 men and 612 women). Anthropometric measures: weight, height and waist circumferenc...

Coupling to the electron Bernstein wave (EBW) via a phased array of waveguides is experimentally investigated in the Madison symmetric torus reversed field pinch (RFP). EBWs promise to provide localized heating and current drive in overdense plasmas such as those in the RFP, provided a technique can be developed to efficiently couple power from an externally launched electromagnetic wave to the...

In this paper, we consider a generalization of the state-of-art discriminative method for optimizing the conditional likelihood in Hidden Markov Models (HMMs), called the Extended Baum-Welch (EBW) algorithm, that has had significant impact on the speech recognition community. We propose a generalized form of EBW update rules that can be associated with a weighted sum of updated and initial mode...