The Experts below are selected from a list of 47691 Experts worldwide ranked by ideXlab platform

Conor Ryan - One of the best experts on this subject based on the ideXlab platform.

James Vincent Patten - One of the best experts on this subject based on the ideXlab platform.

Zhengming Ding - One of the best experts on this subject based on the ideXlab platform.

  • CVPR - Marginalized Latent Semantic Encoder for Zero-Shot Learning
    2019 IEEE CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019
    Co-Authors: Zhengming Ding
    Abstract:

    Zero-shot learning has been well explored to precisely identify new unobserved classes through a visual-Semantic Function obtained from the existing objects. However, there exist two challenging obstacles: one is that the human-annotated Semantics are insufficient to fully describe the visual samples; the other is the domain shift across existing and new classes. In this paper, we attempt to exploit the intrinsic relationship in the Semantic manifold when given Semantics are not enough to describe the visual objects, and enhance the generalization ability of the visual-Semantic Function with marginalized strategy. Specifically, we design a Marginalized Latent Semantic Encoder (MLSE), which is learned on the augmented seen visual features and the latent Semantic representation. Meanwhile, latent Semantics are discovered under an adaptive graph reconstruction scheme based on the provided Semantics. Consequently, our proposed algorithm could enrich visual characteristics from seen classes, and well generalize to unobserved classes. Experimental results on zero-shot benchmarks demonstrate that the proposed model delivers superior performance over the state-of-the-art zero-shot learning approaches.

  • Zero-Shot Learning
    Learning Representation for Multi-View Data Analysis, 2018
    Co-Authors: Zhengming Ding, Handong Zhao, Yun Fu
    Abstract:

    Zero-shot learning targets at precisely recognizing unseen categories through a shared visual-Semantic Function, which is built on the seen categories and expected to well adapt to unseen categories. However, the Semantic gap across visual features and their underlying Semantics is still the most challenging obstacle. In this chapter, we tackle this issue by exploiting the intrinsic relationship in the Semantic manifold and enhancing the transferability of visual-Semantic Function. Specifically, we propose an Adaptive Latent Semantic Representation (ALSR) model in a sparse dictionary learning scheme, where a generic Semantic dictionary is learned to connect the latent Semantic space with visual feature space. To build a fast inference model, we explore a non-linear network to approximate the latent sparse Semantic representation, which lies in the Semantic manifold space. Consequently, our model could extract a variety of visual characteristics within seen classes, which can be well generalized to unobserved classes.

Karalyn Patterson - One of the best experts on this subject based on the ideXlab platform.

  • taking both sides do unilateral anterior temporal lobe lesions disrupt Semantic memory
    Brain, 2010
    Co-Authors: Matthew Lambon A Ralph, Lisa Cipolotti, Facundo Manes, Karalyn Patterson
    Abstract:

    The most selective disorder of central conceptual knowledge arises in Semantic dementia, a degenerative condition associated with bilateral atrophy of the inferior and polar regions of the temporal lobes. Likewise, Semantic impairment in both herpes simplex virus encephalitis and Alzheimer's disease is typically associated with bilateral, anterior temporal pathology. These findings suggest that conceptual representations are supported via an interconnected, bilateral, anterior temporal network and that it may take damage to both sides to produce an unequivocal deficit of central Semantic memory. We tested and supported this hypothesis by investigating a case series of 20 patients with unilateral temporal damage (following vascular accident or resection for tumour or epilepsy), utilizing a test battery that is sensitive to Semantic impairment in Semantic dementia. Only 1/20 of the cases, with a unilateral left lesion, exhibited even a mild impairment on the receptive Semantic measures. On the expressive Semantic tests of naming and fluency, average performance was worse in the left- than right-unilateral cases, but even in this domain, only one left-lesion case had scores consistently more than two standard deviations below control means. These results fit with recent parallel explorations of Semantic Function using repetitive transcranial magnetic stimulation as well as Functional imaging in stroke aphasic and neurologically intact participants. The evidence suggests that both left and right anterior temporal lobe regions contribute to the representation of Semantic memory and together may form a relatively damage-resistant, robust system for this critical aspect of higher cognition.

Samuel C. Rickless - One of the best experts on this subject based on the ideXlab platform.

  • The Semantic Function of chained pronouns
    Analysis, 1998
    Co-Authors: Samuel C. Rickless
    Abstract:

    The pronouns in (1), if used non-demonstratively, are anaphoric on the antecedent indefinite noun phrase 'a man' that heads (1). Now it might be thought that these pronouns Function Semantically as do the bound variables of quantification theory. But recent work in linguistics and in the philosophy of language (see, for example, Evans 1977 and Neale 1990) suggests that this is not so. What, then, is the Semantic Function of these pronouns? In answer to this question, a number of theories have been proposed. In ? 2-4 of this paper, I argue that none of these theories does justice to our intuitions concerning the truth-conditions of the sentences in (1). In the last section, S5, I briefly sketch a theory that accommodates these very intuitions.