The Experts below are selected from a list of 54999 Experts worldwide ranked by ideXlab platform

Ankur Srivastava - One of the best experts on this subject based on the ideXlab platform.

  • Effective techniques for the generalized low-power Binding Problem
    ACM Transactions on Design Automation of Electronic Systems, 2006
    Co-Authors: Azadeh Davoodi, Ankur Srivastava
    Abstract:

    This article proposes two very fast graph theoretic heuristics for the low power Binding Problem given fixed number of resources and multiple architectures for the resources. First, the generalized low power Binding Problem is formulated as an Integer Linear Programming (ILP) Problem that happens to be an NP-complete task to solve. Then two polynomial-time heuristics are proposed that provide a speedup of up to 13.7 with an extremely low penalty for power when compared to the optimal ILP solution for our selected benchmarks.

  • effective graph theoretic techniques for the generalized low power Binding Problem
    International Symposium on Low Power Electronics and Design, 2003
    Co-Authors: Azadeh Davoodi, Ankur Srivastava
    Abstract:

    This paper proposes two very fast graph theoretic heuristics for the low power Binding Problem given fixed number of resources and multiple architectures for the resources. First the generalized low power Binding Problem is formulated as an Integer Linear Programming(ILP) Problem which happens to be an NP-complete task to solve. Then two polynomial-time heuristics are proposed that provide a speedup of up to 13.7 with an extremely low penalty for power when compared to the optimal ILP solution for our selected benchmarks.

  • ISLPED - Effective graph theoretic techniques for the generalized low power Binding Problem
    Proceedings of the 2003 international symposium on Low power electronics and design - ISLPED '03, 2003
    Co-Authors: Azadeh Davoodi, Ankur Srivastava
    Abstract:

    This paper proposes two very fast graph theoretic heuristics for the low power Binding Problem given fixed number of resources and multiple architectures for the resources. First the generalized low power Binding Problem is formulated as an Integer Linear Programming(ILP) Problem which happens to be an NP-complete task to solve. Then two polynomial-time heuristics are proposed that provide a speedup of up to 13.7 with an extremely low penalty for power when compared to the optimal ILP solution for our selected benchmarks.

  • ICCAD - Predictability: definition, ananlysis and optimization
    Proceedings of the 2002 IEEE ACM international conference on Computer-aided design - ICCAD '02, 2002
    Co-Authors: Ankur Srivastava, Majid Sarrafzadeh
    Abstract:

    Predictability is the quantified from of accuracy. We propose a predictability driven design methodology. The novelty lies in defining and using the idea of predictability. In order to illustrate the basic concepts we focus on the low power Binding Problem. The Binding Problem for low power was solved in [3], [5], but in the presence of in-accuracies, their claims of optimality are imprecise. Our experiments show that these inaccuracies could be as high as 33%. Our methodology could improve this unpredictability to as low as 11% with minimal power penalty (7% on average).

  • Effective graph theoretic techniques for the generalized low power Binding Problem [IC high level synthesis]
    Proceedings of the 2003 International Symposium on Low Power Electronics and Design 2003. ISLPED '03., 1
    Co-Authors: Azadeh Davoodi, Ankur Srivastava
    Abstract:

    This paper proposes two very fast graph theoretic heuristics for the low power Binding Problem given a fixed number of resources and multiple architectures for the resources. First, the generalized low power Binding Problem is formulated as an integer linear programming (ILP) Problem which happens to be an NP-complete task to solve. Then two polynomial-time heuristics are proposed that provide a speedup of up to 13.7 with an extremely low penalty for power when compared to the optimal ILP solution for our selected benchmarks.

Anne Treisman - One of the best experts on this subject based on the ideXlab platform.

  • Solutions to the Binding Problem: Review Progress through Controversy Summary and Convergence
    1999
    Co-Authors: Anne Treisman
    Abstract:

    depend on ratios of activity in neurons with different but overlapping tuning. Whenever the perceptual represenThe Binding Problem, or constellation of Problems, con- tations of simultaneously present objects depend on cerns our capacity to integrate information across time, distributed patterns of firing in populations of cells, the space, attributes, and ideas. The goal of research in this risk of superposition ambiguities within the same neural area is to understand how we can respond to relations network will arise, creating a need to identify and to within relevant subsets of the world but not to relations signal which units belong to the same representation. between arbitrarily selected parts or properties. Lan- For any case of Binding, the Binding Problem can guage comprehension and thinking critically depend on actually be dissected into three separable Problems. correct Binding of syntactic and semantic structures. Different theories have focused primarily on one of the

  • The Binding Problem
    Current opinion in neurobiology, 1996
    Co-Authors: Anne Treisman
    Abstract:

    Abstract Perceptual representations depend on distributed neural codes for relaying the parts and properties of objects. Some mechanism is needed to ‘bind’ the information relating to each object and to distinguish it from others. Possible candidates include cells tuned to conjunctions of features, spatial attention, and synchronized firing across separate but interconnected areas of the brain. Deficits in neurological patients suggest a role for the parietal cortex in the Binding process. Several current models combine these ideas.

  • Parietal contributions to visual feature Binding: evidence from a patient with bilateral lesions
    Science (New York N.Y.), 1995
    Co-Authors: Stacia R. Friedman-hill, Lynn C. Robertson, Anne Treisman
    Abstract:

    Neurophysiologists have documented the existence of multiple cortical areas responsive to different visual features. This modular organization has sparked theoretical interest in how the "Binding Problem" is solved. Recent data from a neurological patient (R.M.) with bilateral parietal-occipital lesions demonstrates that the Binding Problem is not just a hypothetical construct; it can be a practical Problem, as rare as the selective inability to perceive motion or color. R.M. miscombines colors and shapes even under free viewing conditions and is unable to judge either relative or absolute visual locations. The evidence suggests that a single explanation--an inadequate spatial representation--can account for R.M.'s spatial judgment and feature-Binding deficits.

  • Modularity and attention: Is the Binding Problem real?
    Visual Cognition, 1995
    Co-Authors: Anne Treisman
    Abstract:

    Abstract Van der Heijden rejects the feature integration theory of visual attention (Treisman, 1988, 1993; Treisman & Gelade, 1980) and proposes instead a theory relating modularity in the visual system to selection for action. His positive proposals about the relations between visual processing, intention, and selection for action are interesting, but I do not believe they are incompatible with my theory. In this paper I will focus on and question his arguments about the Binding Problem and the role of attention in visual perception. Van der Heijden attacks two claims that are fundamental to my theory: (1) the idea that modularity gives rise to a “Binding Problem” (the need to specify which of the features present characterize any particular object), and (2) the more general idea that there are limits to capacity at the level of perceptual processing. I will argue that he is wrong to reject the two claims, that the Binding Problem is a real one, for his model as for others, and that his account is more s...

Majid Sarrafzadeh - One of the best experts on this subject based on the ideXlab platform.

  • ICCAD - Predictability: definition, ananlysis and optimization
    Proceedings of the 2002 IEEE ACM international conference on Computer-aided design - ICCAD '02, 2002
    Co-Authors: Ankur Srivastava, Majid Sarrafzadeh
    Abstract:

    Predictability is the quantified from of accuracy. We propose a predictability driven design methodology. The novelty lies in defining and using the idea of predictability. In order to illustrate the basic concepts we focus on the low power Binding Problem. The Binding Problem for low power was solved in [3], [5], but in the presence of in-accuracies, their claims of optimality are imprecise. Our experiments show that these inaccuracies could be as high as 33%. Our methodology could improve this unpredictability to as low as 11% with minimal power penalty (7% on average).

  • Predictability: definition, analysis and optimization [VLSI design]
    IEEE ACM International Conference on Computer Aided Design 2002. ICCAD 2002., 1
    Co-Authors: Ankur Srivastava, Majid Sarrafzadeh
    Abstract:

    Predictability is the quantified form of accuracy. We propose a predictability driven design methodology. The novelty lies in defining and using the idea of predictability. In order to illustrate the basic concepts we focus on the low power Binding Problem. The Binding Problem for low power was solved in (A. Raghunathan et al, Procs of IEEE Symp. on Ccts. and Systems, 1995), (J.M. Chang et al, Proc. Design Automation Conference, pp. 29-35, 1995), but in the presence of inaccuracies, their claims of optimality are imprecise. Our experiments show that these inaccuracies could be as high as 33%. Our methodology could improve this unpredictability to as low as 11% with minimal power penalty (7% on average).

Robert Desimone - One of the best experts on this subject based on the ideXlab platform.

David Whitney - One of the best experts on this subject based on the ideXlab platform.