Výsledky vyhledávání - "Burns, Thomas A."

Report

Associative memory inspires improvements for in-context learning using a novel attention residual stream architecture

Autor: Burns, Thomas F, Fukai, Tomoki, Earls, Christopher J

Large language models (LLMs) demonstrate an impressive ability to utilise information within the context of their input sequences to appropriately respond to data unseen by the LLM during its training procedure. This ability is known as in-context le

Externí odkaz: http://arxiv.org/abs/2412.15113

Zobrazit plný text záznamu

Report

Generative AI Policy and Governance Considerations for Health Security in Southeast Asia

Autor: Burns, Thomas F

Southeast Asia is a geopolitically and socio-economically significant region with unique challenges and opportunities. Intensifying progress in generative AI against a backdrop of existing health security threats makes applications of AI to mitigate

Externí odkaz: http://arxiv.org/abs/2411.14435

Zobrazit plný text záznamu

Report

Semantically-correlated memories in a dense associative model

Autor: Burns, Thomas F

I introduce a novel associative memory model named Correlated Dense Associative Memory (CDAM), which integrates both auto- and hetero-association in a unified framework for continuous-valued memory patterns. Employing an arbitrary graph structure to

Externí odkaz: http://arxiv.org/abs/2404.07123

Zobrazit plný text záznamu

Report

Simplicial Hopfield networks

Autor: Burns, Thomas F, Fukai, Tomoki

Publikováno v: International Conference on Learning Representations 2023

Hopfield networks are artificial neural networks which store memory patterns on the states of their neurons by choosing recurrent connection weights and update rules such that the energy landscape of the network forms attractors around the memories.

Externí odkaz: http://arxiv.org/abs/2305.05179

Zobrazit plný text záznamu

Report

Efficient, probabilistic analysis of combinatorial neural codes

Autor: Burns, Thomas F, Irwansyah

Artificial and biological neural networks (ANNs and BNNs) can encode inputs in the form of combinations of individual neurons' activities. These combinatorial neural codes present a computational challenge for direct and efficient analysis due to the

Externí odkaz: http://arxiv.org/abs/2210.10492

Zobrazit plný text záznamu

Report

Detecting danger in gridworlds using Gromov's Link Condition

Autor: Burns, Thomas F, Tang, Robert

Gridworlds have been long-utilised in AI research, particularly in reinforcement learning, as they provide simple yet scalable models for many real-world applications such as robot navigation, emergent behaviour, and operations research. We initiate

Externí odkaz: http://arxiv.org/abs/2201.06274

Zobrazit plný text záznamu