NSTL国家科技图书文献中心

1. MATADOR: Automated System-on-Chip Tsetlin Machine Design Generation for Edge Applications NSTL国家科技图书文献中心

Tousif Rahman | Gang Mao ... - 《2024 Design, Automation & Test in Europe Conference & Exhibition: DATE 2024, Valencia, Spain, 25-27 March 2024, [v.1]》 - Design, Automation & Test in Europe Conference & Exhibition - 2024, - 1～6 - 共6页

摘要：System-on-Chip Field-Programmable Gate Arrays (SoC-FPGAs) offer significant throughput gains for machine learning (ML) edge inference applications via the design of co-processor accelerator systems. H...

关键词： Training | Power demand | Scalability | Pipelines | Throughput | Inference algorithms | System-on-chip

2. Logic-Based Machine Learning with Reproducible Decision Model Using the Tsetlin Machine NSTL国家科技图书文献中心

Olga Tarasyuk | Anatoliy Gorbenko ... - 《2023 IEEE 12th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, Vol. 1: IEEE 12th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), 7-9 September 2023, Dortmund, Germany》 - IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications - 2023, - 465～470 - 共6页

摘要：Tsetlin Machine (TM) is a recent automaton-based algorithm for reinforcement learning. It has demonstrated competitive accuracy on many popular benchmarks while providing a natural interpretability. D...

关键词： Productivity | Neurons | Stochastic processes | Artificial neural networks | Reinforcement learning | Benchmark testing | Reproducibility of results

3. Parallel Symbiotic Random Number Generator for Training Tsetlin Machines on FPGA NSTL国家科技图书文献中心

Tousif Rahman | Gang Mao ... - 《2023 International Symposium on the Tsetlin Machine: International Symposium on the Tsetlin Machine (ISTM), 29-30 August 2023, Newcastle upon Tyne, United Kingdom》 - International Symposium on the Tsetlin Machine - 2023, - 1～8 - 共8页

摘要：The logical proposition (clause) based underpinning and computationally minimal training procedure of Tsetlin Machines (TM) allows for exploration of on-chip training on resource limited FPGAs for edg...

关键词： Training | Symbiosis | Real-time systems | Software | System-on-chip | Table lookup | Field programmable gate arrays

4. REDRESS: Generating Compressed Models for Edge Inference Using Tsetlin Machines NSTL国家科技图书文献中心

Sidharth Maheshwari | Tousif Rahman ... - 《IEEE Transactions on Pattern Analysis and Machine Intelligence》 - 2023,45(9) - 11152～11168 - 共17页 - 被引量：2

摘要：Inference at-the-edge using embedded machine learning models is associated with challenging trade-offs between resource metrics, such as energy and memory footprint, and the performance metrics, such ...

关键词： Computational modeling | Training | Learning automata | Artificial neural networks | Random access memory | Runtime | Low-power electronics

5. Introducing TRIM Automata for Tsetlin Machines NSTL国家科技图书文献中心

Sidharth Maheshwari | Tousif Rahman ... - 《2023 International Symposium on the Tsetlin Machine: International Symposium on the Tsetlin Machine (ISTM), 29-30 August 2023, Newcastle upon Tyne, United Kingdom》 - International Symposium on the Tsetlin Machine - 2023, - 1～4 - 共4页

摘要：The learning automaton is the core element of Tsetlin Machines (TM) that is accessed and modified repeatedly over the course of training. It posits a crucial bottleneck to on-chip training as it requi...

关键词： Training | Costs | Learning automata | Memory management | Automata | System-on-chip | Testing

6. Systematic Search for Optimal Hyper-parameters of the Tsetlin Machine on MNIST Dataset NSTL国家科技图书文献中心

Olga Tarasyuk | Tousif Rahman ... - 《2023 International Symposium on the Tsetlin Machine: International Symposium on the Tsetlin Machine (ISTM), 29-30 August 2023, Newcastle upon Tyne, United Kingdom》 - International Symposium on the Tsetlin Machine - 2023, - 1～8 - 共8页

摘要：Modern Machine Learning (ML) models have a significant number of hyper-parameters that need adjusting to leverage performance and energy efficiency for a given model configuration during training. Thi...

关键词： Training | Systematics | Sensitivity | Machine learning | Learning (artificial intelligence) | Energy efficiency | Complexity theory

7. IMBUE: In-Memory Boolean-to-CUrrent Inference ArchitecturE for Tsetlin Machines NSTL国家科技图书文献中心

Omar Ghazal | Simranjeet Singh ... - 《2023 IEEE/ACM International Symposium on Low Power Electronics and Design: ISLPED 2023, Vienna, Austria, 7-8 August 2023》 - IEEE/ACM International Symposium on Low Power Electronics and Design - 2023, - 1～6 - 共6页

摘要：In-memory computing for Machine Learning (ML) applications remedies the von Neumann bottlenecks by organizing computation to exploit parallelism and locality. Non-volatile memory devices such as Resis...

关键词： Performance evaluation | Power demand | Nonvolatile memory | Scalability | Resistive RAM | Computer architecture | Switches

8. MILEAGE: An Automated Optimal Clause Search Paradigm for Tsetlin Machines NSTL国家科技图书文献中心

Tousif Rahman | Sidharth Maheshwari ... - 《2022 International Symposium on the Tsetlin Machine: ISTM 2022, Grimstad, Norway, 20-21 June 2022》 - International Symposium on the Tsetlin Machine - 2022, - 49～52 - 共4页

摘要：The logic based underpinning of the Tsetlin Machine (TM) offers substantial benefits for compute-efficient inference. The caveat to this is the large memory footprint of the TM growing substantially a...

关键词： Training | Degradation | Source coding | Computational modeling | Robustness | Complexity theory | Standards

9. Data Booleanization for Energy Efficient On-Chip Learning using Logic Driven AI NSTL国家科技图书文献中心

Tousif Rahman | Adrian Wheeldon ... - 《2022 International Symposium on the Tsetlin Machine: ISTM 2022, Grimstad, Norway, 20-21 June 2022》 - International Symposium on the Tsetlin Machine - 2022, - 29～36 - 共8页

摘要：The incorporation of Machine Learning (ML) models into smart devices and wearables has increased design focus on scalability and energy efficiency considerations. To transition towards on-device train...

关键词： Training | Wearable computers | Machine learning | Minimization | Energy efficiency | Encoding | Compaction

10. Learning automata based energy-efficient AI hardware design for IoT applications SCOPUS Scopus数据库 EI 工程索引 SCIE Web of Science核心 NSTL国家科技图书文献中心

AdrianWheeldon | Rishad Shafik ... - 《Philosophical transactions of the Royal Society. Mathematical, physical, and engineering sciences》 - 2020,378(2182) - 20190593-1～20190593-18 - 共18页 - 被引量：10

摘要：Energy efficiency continues to be the core design challenge for artificial intelligence (AI) hardware designers. In this paper, we propose a new AI hardware architecture targeting Internet of Things a...

摘要：System-on-Chip Field-Programmable Gate Arrays (SoC-FPGAs) offer significant throughput gains for machine learning (ML) edge inference applications via the design of co-processor accelerator systems. H...

关键词： Training | Power demand | Scalability | Pipelines | Throughput | Inference algorithms | System-on-chip

摘要：Tsetlin Machine (TM) is a recent automaton-based algorithm for reinforcement learning. It has demonstrated competitive accuracy on many popular benchmarks while providing a natural interpretability. D...

关键词： Productivity | Neurons | Stochastic processes | Artificial neural networks | Reinforcement learning | Benchmark testing | Reproducibility of results

摘要：The logical proposition (clause) based underpinning and computationally minimal training procedure of Tsetlin Machines (TM) allows for exploration of on-chip training on resource limited FPGAs for edg...

关键词： Training | Symbiosis | Real-time systems | Software | System-on-chip | Table lookup | Field programmable gate arrays

摘要：Inference at-the-edge using embedded machine learning models is associated with challenging trade-offs between resource metrics, such as energy and memory footprint, and the performance metrics, such ...

关键词： Computational modeling | Training | Learning automata | Artificial neural networks | Random access memory | Runtime | Low-power electronics

摘要：The learning automaton is the core element of Tsetlin Machines (TM) that is accessed and modified repeatedly over the course of training. It posits a crucial bottleneck to on-chip training as it requi...

关键词： Training | Costs | Learning automata | Memory management | Automata | System-on-chip | Testing

摘要：Modern Machine Learning (ML) models have a significant number of hyper-parameters that need adjusting to leverage performance and energy efficiency for a given model configuration during training. Thi...

关键词： Training | Systematics | Sensitivity | Machine learning | Learning (artificial intelligence) | Energy efficiency | Complexity theory

摘要：In-memory computing for Machine Learning (ML) applications remedies the von Neumann bottlenecks by organizing computation to exploit parallelism and locality. Non-volatile memory devices such as Resis...

关键词： Performance evaluation | Power demand | Nonvolatile memory | Scalability | Resistive RAM | Computer architecture | Switches

摘要：The logic based underpinning of the Tsetlin Machine (TM) offers substantial benefits for compute-efficient inference. The caveat to this is the large memory footprint of the TM growing substantially a...

关键词： Training | Degradation | Source coding | Computational modeling | Robustness | Complexity theory | Standards

摘要：The incorporation of Machine Learning (ML) models into smart devices and wearables has increased design focus on scalability and energy efficiency considerations. To transition towards on-device train...

关键词： Training | Wearable computers | Machine learning | Minimization | Energy efficiency | Encoding | Compaction

摘要：Energy efficiency continues to be the core design challenge for artificial intelligence (AI) hardware designers. In this paper, we propose a new AI hardware architecture targeting Internet of Things a...

关键词： Tsetlin machines | neural networks | artificial intelligence hardware design | energy efficiency

4008-161-200 800-990-8900

国家科技图书文献中心