Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 33 for “"Parallel search"”.

  1. The role of crowding in parallel search

    … in the peripheral (Pelli, 2008). In three visual search experiments the contribution of visual crowding to reaction time performance in an efficient search task was evaluated by varying the factors known to affect the strength of crowding: spacing between objects and similarity. Traditionally, …

    uiuc Repository record for The role of crowding in parallel search (opens in a new tab)

  2. Algorithmic skeletons for exact combinatorial search at scale

    Exact combinatorial search is essential to a wide range of application areas including constraint optimisation, graph matching, and computer algebra. Solutions to combinatorial problems are found by systematically exploring a search space, either to enumerate solutions, determine if a specific …

    glasgow Download PDF for Algorithmic skeletons for exact combinatorial search at scale (opens in a new tab)

  3. COMPUTE-IN-MEMORY ON EMERGING MEMORY TECHNOLOGY: FROM DEVICE TO ALGORITHM

    … essential data operations including storage, parallel search, and neural network. Last, I will discuss the conceptualization and demonstration of a programmable parallel search architecture - analog content-addressable memory (ACAM) on complementary Si-CMOS ferroelectric …

    penn Repository record for COMPUTE-IN-MEMORY ON EMERGING MEMORY TECHNOLOGY: FROM DEVICE TO ALGORITHM (opens in a new tab)

  4. EFFICIENT AND AUTOMATED SCALING OF DISTRIBUTED AI INFRASTRUCTURE

    … efficient scaling along all three. On data-parallel training, we co-design numerics and systems to stabilize and accelerate adversarial learning (ParaGAN). On model-parallel training, we cut auto-parallel search time by orders of magnitude through subgraph mining and analytical cost modeling …

    nus Repository record for EFFICIENT AND AUTOMATED SCALING OF DISTRIBUTED AI INFRASTRUCTURE (opens in a new tab) Download file for EFFICIENT AND AUTOMATED SCALING OF DISTRIBUTED AI INFRASTRUCTURE (opens in a new tab)

  5. Parallelism in Constraint Programming

    Writing efficient parallel programs is the biggest challenge of the software industry for the foreseeable future. We are currently in a time when parallel computers are the norm, not the exception. Soon, parallel processors will be standard even in cell phones. Without drastic changes in hardware …

    lund Repository record for Parallelism in Constraint Programming (opens in a new tab) Download PDF for Parallelism in Constraint Programming (opens in a new tab)

  6. Automated Exploration of the ASIC Design Space for Minimum Power-Delay-Area Product at the Register Transfer Level

    … better than the initial baseline solution. The parallel search algorithm exhibited 9x speed up when executed on 10 machines simultaneously. These two new methods also characterize the design space for the given RTL code by generating power-delay-area points in addition to the minimum PDA point …

    utk-theses Repository record for Automated Exploration of the ASIC Design Space for Minimum Power-Delay-Area Product at the Register Transfer Level (opens in a new tab)

  7. Performance of single dwell serial synchronization system

    … acquisition process and a brief description of parallel search process is introduced. The results demonstrate the dependence of the sync }lionization times on the information signal and permit optimal design of the synchronization detector.</p>

    njit Repository record for Performance of single dwell serial synchronization system (opens in a new tab)

  8. Using reinforcement learning in solving exam timetabling problems

    There is increasing interest in the area of research involving automated technologies for solving problems. High generality and ease of reusability have become a major goal in the development of Search methodologies within automation. One search method that meets these requirements is the …

    qu-belfast Repository record for Using reinforcement learning in solving exam timetabling problems (opens in a new tab) Download PDF for Using reinforcement learning in solving exam timetabling problems (opens in a new tab)

  9. 3CAM - a Novel Ternary Content Addressable Memory Using Three-valued Logic

    … (CMOS) storage element that allows for a parallel search operation on a memory stack in addition to the read and write operations yielded by a conventional SRAM storage array. In practice, it is often desirable to be able to store a “don’t care” state for faster searching operation. …

    bucknell Repository record for 3CAM - a Novel Ternary Content Addressable Memory Using Three-valued Logic (opens in a new tab)

  10. Layerwise symbolic knowledge extraction from deep neural networks

    … extraction algorithm. The second algorithm is a parallel search for optimal M-of-N rules that implements a hyperparameter that controls the complexity of the extracted rules. We apply this algorithm to a variety of deep networks and find that although differences in architecture, dataset, and …

    city-london Download PDF for Layerwise symbolic knowledge extraction from deep neural networks (opens in a new tab)

  11. The analysis of slant-from-texture in early vision

    A considerable amount of research exists on the subjective perception of three-dimensional structure from texture gradients. The present set of experiments extends these tests of phenomenal perception by examining the underlying processes used in interpreting slant-from-texture. The first two …

    ubc Repository record for The analysis of slant-from-texture in early vision (opens in a new tab) Download PDF for The analysis of slant-from-texture in early vision (opens in a new tab)

  12. Predictive Parallelization: A Framework for Reducing Tail Latencies of Web Search Queries

    We have become dependent on web search in our everyday lives. Web search services aim to provide fast responses to user queries, making the tail latency more important to reduce than the average latency. With modern multicore servers, intra-query parallelization becomes a desirable technique for …

    rice Repository record for Predictive Parallelization: A Framework for Reducing Tail Latencies of Web Search Queries (opens in a new tab)

  13. Smart Memory: An Inexact Content-Addressable Memory

    … Memory (CAM) is to efficiently search the information stored in the memory, by using hardware rather than software with a corresponding improvement in searching speed. This hardware allows a parallel search by matching the data stored in memory to a search key rather than …

    portland-state Repository record for Smart Memory: An Inexact Content-Addressable Memory (opens in a new tab)

  14. Evolutionary Optimization Algorithms for Nonlinear Systems

    … lead the vectors to get trapped in local minima. Parallel search techniques have more exploratory power. So, depending on the application, different suitable techniques are used. This thesis mainly concentrates on benchmarking three popular algorithms: Real valued Genetic Algorithm (RGA), Particle …

    utahstate Repository record for Evolutionary Optimization Algorithms for Nonlinear Systems (opens in a new tab)

  15. A Low Power Beacon for Transmitting Location Data via the Globalstar Satellite System

    … results. The simulated correlator employed a parallel search strategy and used a truncated version of the Globalstar system code sequence. Finally, a threshold analysis was performed to determine if better receiver performance could be obtained.

    vt Repository record for A Low Power Beacon for Transmitting Location Data via the Globalstar Satellite System (opens in a new tab)

  16. Algorithm Based Fault Tolerance: A Perspective from Algorithmic and Communication Characteristics of Parallel Algorithms

    … algorithms for linear systems has been the research focus of ABFT schemes. As a consequence, it necessitates a comprehensive exploration of ABFT research to widen its scope to other types of parallel algorithms and applications. In this thesis, we go beyond traditional ABFT and focus on other …

    concordia Download PDF for Algorithm Based Fault Tolerance: A Perspective from Algorithmic and Communication Characteristics of Parallel Algorithms (opens in a new tab)

  17. A Multi-Platform High-Performance Computing Framework for Combinatorial Optimization

    … a time and compute-intensive task. Current research has focused on designing accelerators for solving problems in the NP-complete Quadratic Unconstrained Binary Optimization (QUBO) format due to its simplicity. However, these accelerators are difficult to scale and are inefficient at solving …

    toronto-retro Repository record for A Multi-Platform High-Performance Computing Framework for Combinatorial Optimization (opens in a new tab)

  18. Tuning Parallel Applications in Parallel

    … Our system, Active Harmony, takes a search-based collaborative approach to auto-tuning. Application programmers, library writers and compilers collaborate to describe and export a set of performance related tunable parameters to the Active Harmony system. These parameters define a …

    maryland Repository record for Tuning Parallel Applications in Parallel (opens in a new tab)

Page 1 of 2