Attending to a particular feature modulates the activity of neurons throughout the visual field with the result that relevant features are enhanced while irrelevant features are suppressed. Do these modulatory influences merely lead to a gating of relevant features, or does attention have a direct impact on the representation of feature space, leading to a different percept depending on the con...

A fundamental problem in machine learning is understanding the conditions for which a learning algorithm works well. Understanding an algorithm's strengths and weaknesses and being able to compare two algorithms with each other are necessary for designers to develop (or select) learning algorithms for a speci c problem. Generally, one can attempt to analyze and understand algorithms either theo...

This paper presents a significant modification to the AdaSS (Adaptive Splitting and Selection) algorithm, which was developed several years ago. The method is based on the simultaneous partitioning of the feature space and an assignment of a compound classifier to each of the subsets. The original version of the algorithm uses a classifier committee and a majority voting rule to arrive at a dec...

Feature detectors using a quadratic nonlinearity in the ltering stage are known to have some advantages over linear detectors; here we consider how their scale-space properties compare. In particular, we investigate the question whether, like linear detectors, quadratic feature detectors permit a scale-selection scheme with the \causality property", which guarantees that features are never crea...

Modelling varying speaking style remains a challenge to state of the art speech recognition and synthesis systems. Vowel and consonant reduction have been identified as correlative to speaking style variation, but still lack a common measurement. The reduction phenomena are often observed without consideration of coarticulation and assimilation effects, and as a result of speaking rate variabil...

We present a novel fingerprint alignment and matching scheme that utilizes ridge feature maps to represent, align and match fingerprint images. The technique described here obviates the need for extracting minutiae points or the core point to either align or match fingerprint images. The proposed scheme examines the ridge strength (in local neighborhoods of the fingerprint image) at various ori...

The infamous exploration-exploitation dilemma is one of the oldest and most important problems in reinforcement learning (RL). Deliberate and effective exploration is necessary for RL agents to succeed in most environments. However, until very recently even very sophisticated RL algorithms employed simple, undirected exploration strategies in large-scale RL tasks. We introduce a new optimistic ...

When presented with a yellow Volkswagen and a red Ferrari, how does the brain figure out which color goes with which car? The binding problem refers to how the visual system pre-consciously combines visual features of objects in the physical world to create coherent mental equivalents in our consciousness. I discuss why feature binding is a problem for our brains despite its seemingly effortles...

We extend the well-known technique of constrained Maximum Likelihood Linear Regression (MLLR) to compute a projection (instead of a full rank transformation) on the feature vectors of the adaptation data. We model the projected features with phone-dependent Gaussian distributions and also model the complement of the projected space with a single class-independent, speaker-specific Gaussian dist...

Speaker adaptation is critical for modern speech recognition systems. Due to the computational and multi-channel model sharing considerations, the use of model adaptation techniques is limited in telephony speech recognition systems. On the other hand, feature space adaptation methods such as feature space maximum likelihood linear regression (fMLLR) are efficient approaches suitable for teleph...