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 19 of 19 for “"Probabilistic Computing"”.

  1. Accelerating Probabilistic Computing with a Stochastic Processing Unit

    … learning becomes a more important workload for computing systems than ever before. Probabilistic computing is a popular approach in statistical machine learning, which solves problems by iteratively generating samples from parameterized distributions. As an alternative to Deep Neural Networks, …

    duke Repository record for Accelerating Probabilistic Computing with a Stochastic Processing Unit (opens in a new tab)

  2. New materials for magnetic tunnel junction devices for neuromorphic and probabilistic computing

    … showing great potential to achieve beyond CMOS computing tasks. Progress in materials has greatly improved the high tunnel magnetoresistance within these junctions, making advances in neuromorphic and probabilistic computing architectures, designed towards artificial intelligence and big data …

    texas Repository record for New materials for magnetic tunnel junction devices for neuromorphic and probabilistic computing (opens in a new tab)

  3. Algorithm-Hardware Co-design for Deep Learning and Probabilistic Computing with Compute-in-Memory Accelerators

    … resources. The CIM scheme is further extended to probabilistic computing, where the memory and circuit intrinsic stochasticity are no longer harmful to model accuracy but provides energy-efficient random number generation. A novel CIM accelerator for Bayesian neural network is proposed to generate …

    gatech Repository record for Algorithm-Hardware Co-design for Deep Learning and Probabilistic Computing with Compute-in-Memory Accelerators (opens in a new tab)

  4. Harnessing resilience: biased voltage overscaling for probabilistic signal processing

    A central component of modern computing is the idea that computation requires determinism. Contrary to this belief, the primary contribution of this work shows that useful computation can be accomplished in an error-prone fashion. Focusing on low-power computing and the increasing push toward …

    gatech Repository record for Harnessing resilience: biased voltage overscaling for probabilistic signal processing (opens in a new tab)

  5. Probabilistic Proof-carrying Code

    … of the program. Realistic implementations of computing systems are susceptible to probabilistic behaviours that can alter the execution of a program in ways that can result in corruption or security breaches. We investigate the use of a probabilistic bytecode language to model deterministic …

    ottawa-retro Repository record for Probabilistic Proof-carrying Code (opens in a new tab)

  6. MAGNETIC TUNNEL JUNCTION BASED TRUE RANDOM NUMBER GENERATOR AND PROBABILISTIC ISING MACHINE

    … conventional CMOS technology has powered modern computing for decades, but it is now increasingly limited by physical and economic constraints. Meanwhile, computational demand continues to grow rapidly, especially for artificial intelligence, security, and large-scale optimization. This motivates …

    nus Repository record for MAGNETIC TUNNEL JUNCTION BASED TRUE RANDOM NUMBER GENERATOR AND PROBABILISTIC ISING MACHINE (opens in a new tab) Download file for MAGNETIC TUNNEL JUNCTION BASED TRUE RANDOM NUMBER GENERATOR AND PROBABILISTIC ISING MACHINE (opens in a new tab)

  7. STRAINTRONIC NANOMAGNETIC DEVICES FOR NON-BOOLEAN COMPUTING

    … very attractive for both Boolean and non-Boolean computing applications. There was relatively little research on straintronic switching in devices built with real nanomagnets that invariably have defects and imperfections, or their adaptation to non-Boolean computing, both of which have been …

    vcu Repository record for STRAINTRONIC NANOMAGNETIC DEVICES FOR NON-BOOLEAN COMPUTING (opens in a new tab)

  8. Generative Modeling for Inverse Problems in Materials Science

    … both conditional and unconditional, allowing for probabilistic techniques to operate in a new, previously intractable, regime. A wide array of scientific pursuits are interested in probabilistic techniques due to their ability to capture uncertainty, regularize ill-posed problems through the …

    gatech Repository record for Generative Modeling for Inverse Problems in Materials Science (opens in a new tab)

  9. Fabrication and Simulation of Nanomagnetic Devices for Information Processing

    … that dissipates energy as low as 7.1 fJ. Next, probabilistic (p-) bits implemented with low energy barrier nanomagnets (LBMs) are also analyzed through simulations, and plots show that the probability curves are not affected much by reasonable variations in either thickness or lateral dimensions …

    vcu Repository record for Fabrication and Simulation of Nanomagnetic Devices for Information Processing (opens in a new tab)

  10. Resistive-Switching Materials: From Fundamental Transport to Neuromorphic Applications

    The growing demands on modern computing hardware and the limits of conventional MOS scaling have renewed interest in materials that can support energy-efficient and reliable electronic operation. Materials that exhibit electrically driven phase transitions offer such functionality. In these …

    buffalo Repository record for Resistive-Switching Materials: From Fundamental Transport to Neuromorphic Applications (opens in a new tab)

  11. Probabilistic-Based Computing Transformation with Reconfigurable Logic Fabrics

    … explosion requires a huge quantum leap in our computing power and energy efficiency. However, with the Moore's law dwindling quickly, the physical limits of CMOS technology make it almost intractable to achieve high energy efficiency if the traditional "deterministic and precise" computing …

    ucf Repository record for Probabilistic-Based Computing Transformation with Reconfigurable Logic Fabrics (opens in a new tab)

  12. Magnetic Tunnel Junctions For Next Generation Of Conventional And Unconventional Computing Schemes

    … promising devices for next generation memory and computing technologies. These features include multifunctionality, tunable stochasticity, and capability of being tuned by multiple forces. In this dissertation, I demonstrate how these properties make MTJs promising solutions in conventional memory …

    umn Repository record for Magnetic Tunnel Junctions For Next Generation Of Conventional And Unconventional Computing Schemes (opens in a new tab)

  13. MAGNETIC TUNNEL JUNCTION-BASED STOCHASTIC HARDWARE FOR NEURAL AND ARTIFICIAL INTELLIGENCE SYSTEMS

    … hardware prototypes, exploiting their intrinsic probabilistic switching to realize both nonlinearity and randomness. First, STT-MTJs are used to implement sigmoidal activation in a proof-of-concept neural network for speech separation, showing promising potential for reductions in energy and area …

    nus Repository record for MAGNETIC TUNNEL JUNCTION-BASED STOCHASTIC HARDWARE FOR NEURAL AND ARTIFICIAL INTELLIGENCE SYSTEMS (opens in a new tab) Download file for MAGNETIC TUNNEL JUNCTION-BASED STOCHASTIC HARDWARE FOR NEURAL AND ARTIFICIAL INTELLIGENCE SYSTEMS (opens in a new tab)

  14. Probabilistic Knowledge-guided Machine Learning in Engineering and Geoscience Systems

    … This thesis proposes the development of novel Probabilistic Knowledge-Guided Machine Learning (P-KGML) models to address these concerns. P-KGML models integrate domain knowledge and probabilistic reasoning to improve the explainability, generalization, and physical consistency of ML outputs. …

    umn Repository record for Probabilistic Knowledge-guided Machine Learning in Engineering and Geoscience Systems (opens in a new tab)

  15. Hardware-Software Co-Optimization for Bayesian Neural Networks and Spiking Transformers

    … design paradigms – brain-inspired neuromorphic computing ideas such as In-Memory Computing (IMC) and Spiking Neural Network (SNN), emerging compute elements such as Non-Volatile Memory (NVM), and energy-efficient computing methodologies such as probabilistic or stochastic computing. At the same …

    kings Repository record for Hardware-Software Co-Optimization for Bayesian Neural Networks and Spiking Transformers (opens in a new tab) Download PDF for Hardware-Software Co-Optimization for Bayesian Neural Networks and Spiking Transformers (opens in a new tab)

  16. Learning, probabilistic, and asynchronous technologies for an ultra efficient datapath

    … updated with floating gate technology. Probabilistic digital signal processing is also presented where the probabilistic operation is due to the statistical delay through the asynchronous circuits. Results show successful image processing with probabilistic operation in the least …

    gatech Repository record for Learning, probabilistic, and asynchronous technologies for an ultra efficient datapath (opens in a new tab)

  17. EMERGENT PHENOMENA IN CONDENSED MATTER, SOFT MATTER AND COMPLEX SYSTEMS

    … the improvement of the performance of standard computing architectures encouraged the quest for different approaches aiming at reproducing the computational capability and energy efficiency of the human brain, by mimicking neurons and synapses as probabilistic computing units [Markovic et al., …

    milano Repository record for EMERGENT PHENOMENA IN CONDENSED MATTER, SOFT MATTER AND COMPLEX SYSTEMS (opens in a new tab)