-gram models are the most widely used language models in large vocabulary continuous speech recognition. Since the size of the model grows rapidly with respect to the model order and available training data, many methods have been proposed for pruning the least relevant -grams from the model. However, correct smoothing of the -gram probability distributions is important and performance may degr...

Classification is important problem in data mining. Given a data set, classifier generates meaningful description for each class. Decision trees are most effective and widely used classification methods. There are several algorithms for induction of decision trees. These trees are first induced and then prune subtrees with subsequent pruning phase to improve accuracy and prevent overfitting. In...

Discriminative training schemes, such as Maximum Mutual Information Estimation (MMIE), have been used to improve the accuracy of speech recognition systems trained using Maximum Likelihood Estimation (MLE). In this paper, we present the implementation details of MMIE training in SphinxTrain and baseline results for MMIE training on the Wall Street Journal (WSJ) SI84 and SI284 data sets. This pa...

Abstract Neural network pruning offers great prospects for facilitating the deployment of deep neural networks on computational resource limited devices. architecture search (NAS) provides an efficient way to automatically seek appropriate design compressed model. It is observed that, existing NAS-based methods, there usually a lack layer information when searching optimal architecture. In this...

Partially Observable Markov Decision Processes (POMDPs) are powerful models for planning under uncertainty in partially observable domains. However, computing optimal solutions for POMDPs is challenging because of the high computational requirements of POMDP solution algorithms. Several algorithms use a subroutine to prune dominated vectors in value functions, which requires a large number of l...

Angiogenesis produces primitive vascular networks that need pruning to yield hierarchically organized and functional vessels. Despite the critical importance of vessel pruning to vessel patterning and function, the mechanisms regulating this process are not clear. Here we show that EphrinB2, a well-known player in angiogenesis, is an essential regulator of endothelial cell death and vessel prun...

Induction of decision trees is one of the most successful approaches to supervised machine learning. Branching programs are a generalization of decision trees and, by the boosting analysis, exponentially more eeciently learnable than decision trees. In experiments this advantage has not been seen to materialize. Decision trees are easy to simplify using pruning. For branching programs no such a...

Pruning, a process by which neurons selectively remove exuberant or unnecessary processes without causing cell death, is crucial for the establishment of mature neural circuits during animal development. Yet relatively little is known about molecular and cellular mechanisms that govern neuronal pruning. Holometabolous insects, such as Drosophila, undergo complete metamorphosis and their larval ...

Developmental axon pruning is essential for wiring the mature nervous system, but its regulation remains poorly understood. Here we show that the endosomal-lysosomal pathway regulates developmental pruning of Drosophila mushroom body γ neurons. We demonstrate that the UV radiation resistance-associated gene (UVRAG) functions together with all core components of the phosphatidylinositol 3-kinase...

When approaching a novel visual recognition problem in a specialized image domain, a common strategy is to start with a pre-trained deep neural network and fine-tune it to the specialized domain. If the target domain covers a smaller visual space than the source domain used for pre-training (e.g. ImageNet), the fine-tuned network is likely to be overparameterized. However, applying network prun...