Efficient brain-inspired LLM algorithm/hardware design
Applications are now CLOSED
Overview
This PhD project is aimed at developing a brain-inspired large language model (LLM) system, specifically designed for edge computing applications. This project capitalizes on the principles of neuromorphic computing, which seeks to replicate the human brain's neural architecture, thereby enabling the system to process complex signals with high efficiency and low power consumption in real time. Neuromorphic computing's potential to dramatically enhance the computational speed and power efficiency makes it an ideal foundation for deploying advanced LLMs in environments where rapid response and processing agility are crucial. These LLMs, which have recently achieved significant milestones in natural language understanding and generation, will be tailored to leverage brain-like processing patterns to improve both the speed and accuracy of decision-making processes in AI systems. By combining efficient LLM algorithms with innovative neuromorphic hardware design, this project opens up new possibilities for developing AI systems that are not only more responsive and energy-efficient but also capable of handling the increasing demands of real-world applications like interactive speech systems and autonomous decision-making platforms.
This PhD project is dedicated to constructing a highly efficient, brain-inspired large language model (LLM) system, with a particular focus on developing both the neuromorphic hardware and the LLM algorithms that are optimized for edge computing applications. By drawing inspiration from the human brain's neural structure, the project aims to significantly advance the capabilities of natural language processing by creating neuromorphic chips designed to mimic synaptic connectivity and neural plasticity. These chips will enhance the learning and memory processes of LLMs, enabling faster and more power-efficient processing. Simultaneously, the LLM algorithms will be refined and tailored to leverage the unique characteristics of this brain-like hardware, focusing on reducing latency and maximizing computational efficiency. The integration of these technologies will result in a pioneering system specifically designed for real-time, responsive tasks such as speech recognition and interactive assistance. Comprehensive testing and performance evaluations in real-world scenarios will ensure the system's practical viability and adherence to ethical standards, establishing a new benchmark for the deployment of AI systems that are both technologically innovative and closely aligned with natural human cognitive processes.
[1] Wang, Zhehui, et al. "Enabling Energy-Efficient Deployment of Large Language Models on Memristor Crossbar: A Synergy of Large and Small." IEEE Transactions on Pattern Analysis and Machine Intelligence (2024).
[2] Yao, Man, et al. "Spike-driven transformer." Advances in neural information processing systems 36 (2024).
[3] Zhu, Rui-Jie, et al. "Spikegpt: Generative pre-trained language model with spiking neural networks." arXiv preprint arXiv:2302.13939 (2023).
Funding Information
To be eligible for consideration for a Home DfE or EPSRC Studentship (covering tuition fees and maintenance stipend of approx. £19,237 per annum), a candidate must satisfy all the eligibility criteria based on nationality, residency and academic qualifications.
To be classed as a Home student, candidates must meet the following criteria and the associated residency requirements:
• Be a UK National,
or • Have settled status,
or • Have pre-settled status,
or • Have indefinite leave to remain or enter the UK.
Candidates from ROI may also qualify for Home student funding.
Previous PhD study MAY make you ineligible to be considered for funding.
Please note that other terms and conditions also apply.
Please note that any available PhD studentships will be allocated on a competitive basis across a number of projects currently being advertised by the School.
A small number of international awards will be available for allocation across the School. An international award is not guaranteed to be available for this project, and competition across the School for these awards will be highly competitive.
Academic Requirements:
The minimum academic requirement for admission is normally an Upper Second Class Honours degree from a UK or ROI Higher Education provider in a relevant discipline, or an equivalent qualification acceptable to the University.
Project Summary
Dr Xinming Shi
Full-time: 3 or 3.5 years