Search

Your search keyword '"Tully, Philip"' showing total 20 results
Start Over Author "Tully, Philip" Search Limiters Available in Library Collection
20 results on '"Tully, Philip"'

1. Transformers for End-to-End InfoSec Tasks: A Feasibility Study

Author

Rudd, Ethan M., Rahman, Mohammad Saidur, and Tully, Philip

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Cryptography and Security
Abstract

In this paper, we assess the viability of transformer models in end-to-end InfoSec settings, in which no intermediate feature representations or processing steps occur outside the model. We implement transformer models for two distinct InfoSec data formats - specifically URLs and PE files - in a novel end-to-end approach, and explore a variety of architectural designs, training regimes, and experimental settings to determine the ingredients necessary for performant detection models. We show that in contrast to conventional transformers trained on more standard NLP-related tasks, our URL transformer model requires a different training approach to reach high performance levels. Specifically, we show that 1) pre-training on a massive corpus of unlabeled URL data for an auto-regressive task does not readily transfer to binary classification of malicious or benign URLs, but 2) that using an auxiliary auto-regressive loss improves performance when training from scratch. We introduce a method for mixed objective optimization, which dynamically balances contributions from both loss terms so that neither one of them dominates. We show that this method yields quantitative evaluation metrics comparable to that of several top-performing benchmark classifiers. Unlike URLs, binary executables contain longer and more distributed sequences of information-rich bytes. To accommodate such lengthy byte sequences, we introduce additional context length into the transformer by providing its self-attention layers with an adaptive span similar to Sukhbaatar et al. We demonstrate that this approach performs comparably to well-established malware detection models on benchmark PE file datasets, but also point out the need for further exploration into model improvements in scalability and compute efficiency., Comment: Post-print of a manuscript accepted to ACM Asia-CCS Workshop on Robust Malware Analysis (WoRMA) 2022. 11 Pages total. arXiv admin note: substantial text overlap with arXiv:2011.03040

Published
2022
Full Text
View/download PDF

2. Generative Models for Spear Phishing Posts on Social Media

Author

Seymour, John and Tully, Philip

Subjects
Computer Science - Cryptography and Security, Computer Science - Computers and Society, Computer Science - Learning, Statistics - Machine Learning
Abstract

Historically, machine learning in computer security has prioritized defense: think intrusion detection systems, malware classification, and botnet traffic identification. Offense can benefit from data just as well. Social networks, with their access to extensive personal data, bot-friendly APIs, colloquial syntax, and prevalence of shortened links, are the perfect venues for spreading machine-generated malicious content. We aim to discover what capabilities an adversary might utilize in such a domain. We present a long short-term memory (LSTM) neural network that learns to socially engineer specific users into clicking on deceptive URLs. The model is trained with word vector representations of social media posts, and in order to make a click-through more likely, it is dynamically seeded with topics extracted from the target's timeline. We augment the model with clustering to triage high value targets based on their level of social engagement, and measure success of the LSTM's phishing expedition using click-rates of IP-tracked links. We achieve state of the art success rates, tripling those of historic email attack campaigns, and outperform humans manually performing the same task., Comment: Presented at NIPS Workshop on Machine Deception (2017), 4 page limit plus references, 2 figures

Published
2018

3. ASSOCIATION BETWEEN BLOOD PRESSURE VARIABILITY WITH DEMENTIA AND COGNITIVE IMPAIRMENT: A SYSTEMATIC REVIEW AND META-ANALYSIS

Author

de Heus, RianRianne, Tzourio, Christophe, Lee, Emily, Opozda, Melissa, Vincent, Andrew, Anstey, Kaarin, Hofman, Albert, Kario, Kazuomi, Lattanzi, Simona, Launer, Lenore, Ma, Yuan, Mahajan, Rajiv, Mooijaart, Simon, Nagai, Michiaki, Peters, Ruth, Turnbull, Deborah, Yano, Yuichiro, Claa, Jurgen, ssen, Claassen, Jurgen, and Tully, Philip

Published
2024
Full Text
View/download PDF

4. Spike-Based Bayesian-Hebbian Learning in Cortical and Subcortical Microcircuits

Author

Tully, Philip

Subjects
reinforcement learning, Datorsystem, Computer Systems, Bayes' rule, synaptic plasticity and memory modeling, spiking neural networks, intrinsic excitability, naïve Bayes classifier, Hebbian learning, neuromorphic engineering, temporal sequence learning, attractor network
Abstract

Cortical and subcortical microcircuits are continuously modified throughout life. Despite ongoing changes these networks stubbornly maintain their functions, which persist although destabilizing synaptic and nonsynaptic mechanisms should ostensibly propel them towards runaway excitation or quiescence. What dynamical phenomena exist to act together to balance such learning with information processing? What types of activity patterns do they underpin, and how do these patterns relate to our perceptual experiences? What enables learning and memory operations to occur despite such massive and constant neural reorganization? Progress towards answering many of these questions can be pursued through large-scale neuronal simulations. In this thesis, a Hebbian learning rule for spiking neurons inspired by statistical inference is introduced. The spike-based version of the Bayesian Confidence Propagation Neural Network (BCPNN) learning rule involves changes in both synaptic strengths and intrinsic neuronal currents. The model is motivated by molecular cascades whose functional outcomes are mapped onto biological mechanisms such as Hebbian and homeostatic plasticity, neuromodulation, and intrinsic excitability. Temporally interacting memory traces enable spike-timing dependence, a stable learning regime that remains competitive, postsynaptic activity regulation, spike-based reinforcement learning and intrinsic graded persistent firing levels. The thesis seeks to demonstrate how multiple interacting plasticity mechanisms can coordinate reinforcement, auto- and hetero-associative learning within large-scale, spiking, plastic neuronal networks. Spiking neural networks can represent information in the form of probability distributions, and a biophysical realization of Bayesian computation can help reconcile disparate experimental observations. QC 20170421

Published
2017

5. Large-scale simulations of plastic neural networks on neuromorphic hardware

Author

Knight, James Courtney, Tully, Philip Joseph, Kaplan, Bernhard A, Lansner, Anders, and Furber, Stephen

Subjects
SpiNNaker, learning, plasticity, digital neuromorphic hardware, Bayesian confidence propagation neural network (BCPNN), event-driven simulation, fixed-point accuracy
Abstract

SpiNNaker is a digital, neuromorphic architecture designed for simulating large-scale spiking neural networks at speeds close to biological real-time. Rather than using bespoke analog or digital hardware, the basic computational unit of a SpiNNaker system is a general-purpose ARM processor, allowing it to be programmed to simulate a wide variety of neuron and synapse models. This flexibility is particularly valuable in the study of biological plasticity phenomena. A recently proposed learning rule based on the Bayesian Confidence Propagation Neural Network (BCPNN) paradigm offers a generic framework for modeling the interaction of different plasticity mechanisms using spiking neurons. However, it can be computationally expensive to simulate large networks with BCPNN learning since it requires multiple state variables for each synapse, each of which needs to be updated every simulation time-step. We discuss the trade-offs in efficiency and accuracy involved in developing an event-based BCPNN implementation for SpiNNaker based on an analytical solution to the BCPNN equations, and detail the steps taken to fit this within the limited computational and memory resources of the SpiNNaker architecture. We demonstrate this learning rule by learning temporal sequences of neural activity within a recurrent attractor network which we simulate at scales of up to 2.0 x 104 neurons and 5.1 x 107 plastic synapses: the largest plastic neural network ever to be simulated on neuromorphic hardware. We also run a comparable simulation on a Cray XC-30 supercomputer system and find that, if it is to match the run-time of our SpiNNaker simulation, the super computer system uses approximately 45x more power. This suggests that cheaper, more power efficient neuromorphic systems are becoming useful discovery tools in the study of plasticity in large-scale brain models.

Published
2016
Full Text
View/download PDF

6. Large-Scale Simulations of Plastic Neural Networks on Neuromorphic Hardware

Author

Knight, James C., Tully, Philip J., Kaplan, Bernhard A., Lansner, Anders, Furber, Steve B., Knight, James C., Tully, Philip J., Kaplan, Bernhard A., Lansner, Anders, and Furber, Steve B.

Abstract

SpiNNaker is a digital, neuromorphic architecture designed for simulating large-scale spiking neural networks at speeds close to biological real-time. Rather than using bespoke analog or digital hardware, the basic computational unit of a SpiNNaker system is a general-purpose ARM processor, allowing it to be programmed to simulate a wide variety of neuron and synapse models. This flexibility is particularly valuable in the study of biological plasticity phenomena. A recently proposed learning rule based on the Bayesian Confidence Propagation Neural Network (BCPNN) paradigm offers a generic framework for modeling the interaction of different plasticity mechanisms using spiking neurons. However, it can be computationally expensive to simulate large networks with BCPNN learning since it requires multiple state variables for each synapse, each of which needs to be updated every simulation time-step. We discuss the trade-offs in efficiency and accuracy involved in developing an event-based BCPNN implementation for SpiNNaker based on an analytical solution to the BCPNN equations, and detail the steps taken to fit this within the limited computational and memory resources of the SpiNNaker architecture. We demonstrate this learning rule by learning temporal sequences of neural activity within a recurrent attractor network which we simulate at scales of up to 2.0 x 10(4) neurons and 5.1 x 10(7) plastic synapses: the largest plastic neural network ever to be simulated on neuromorphic hardware. We also run a comparable simulation on a Cray XC-30 supercomputer system and find that, if it is to match the run-time of our SpiNNaker simulation, the super computer system uses approximately 45x more power. This suggests that cheaper, more power efficient neuromorphic systems are becoming useful discovery tools in the study of plasticity in large-scale brain models., QC 20160509

Published
2016
Full Text
View/download PDF

7. Spike-Based Bayesian-Hebbian Learning of Temporal Sequences

Author

Tully, Philip J., Lindén, Henrik, Hennig, Matthias H., Lansner, Anders, Tully, Philip J., Lindén, Henrik, Hennig, Matthias H., and Lansner, Anders

Abstract

Many cognitive and motor functions are enabled by the temporal representation and processing of stimuli, but it remains an open issue how neocortical microcircuits can reliably encode and replay such sequences of information. To better understand this, a modular attractor memory network is proposed in which meta-stable sequential attractor transitions are learned through changes to synaptic weights and intrinsic excitabilities via the spike-based Bayesian Confidence Propagation Neural Network (BCPNN) learning rule. We find that the formation of distributed memories, embodied by increased periods of firing in pools of excitatory neurons, together with asymmetrical associations between these distinct network states, can be acquired through plasticity. The model's feasibility is demonstrated using simulations of adaptive exponential integrate-and-fire model neurons (AdEx). We show that the learning and speed of sequence replay depends on a confluence of biophysically relevant parameters including stimulus duration, level of background noise, ratio of synaptic currents, and strengths of short-term depression and adaptation. Moreover, sequence elements are shown to flexibly participate multiple times in the sequence, suggesting that spiking attractor networks of this type can support an efficient combinatorial code. The model provides a principled approach towards understanding how multiple interacting plasticity mechanisms can coordinate hetero-associative learning in unison., QC 20160817

Published
2016
Full Text
View/download PDF

8. Functional Relevance of Different Basal Ganglia Pathways Investigated in a Spiking Model with Reward Dependent Plasticity

Author

Berthet, Pierre, Lindahl, Mikael, Tully, Philip J., Hellgren-Kotaleski, Jeanette, Lansner, Anders, Berthet, Pierre, Lindahl, Mikael, Tully, Philip J., Hellgren-Kotaleski, Jeanette, and Lansner, Anders

Abstract

The brain enables animals to behaviorally adapt in order to survive in a complex and dynamic environment, but how reward-oriented behaviors are achieved and computed by its underlying neural circuitry is an open question. To address this concern, we have developed a spiking model of the basal ganglia (BG) that learns to dis-inhibit the action leading to a reward despite ongoing changes in the reward schedule. The architecture of the network features the two pathways commonly described in BG, the direct (denoted D1) and the indirect (denoted D2) pathway, as well as a loop involving striatum and the dopaminergic system. The activity of these dopaminergic neurons conveys the reward prediction error (RPE), which determines the magnitude of synaptic plasticity within the different pathways. All plastic connections implement a versatile four-factor learning rule derived from Bayesian inference that depends upon pre- and post-synaptic activity, receptor type, and dopamine level. Synaptic weight updates occur in the D1 or D2 pathways depending on the sign of the RPE, and an efference copy informs upstream nuclei about the action selected. We demonstrate successful performance of the system in a multiple-choice learning task with a transiently changing reward schedule. We simulate lesioning of the various pathways and show that a condition without the D2 pathway fares worse than one without D1. Additionally, we simulate the degeneration observed in Parkinson's disease (PD) by decreasing the number of dopaminergic neurons during learning. The results suggest that the D1 pathway impairment in PD might have been overlooked. Furthermore, an analysis of the alterations in the synaptic weights shows that using the absolute reward value instead of the RPE leads to a larger change in D1., QC 20160815

Published
2016
Full Text
View/download PDF

9. Spike-Based Bayesian-Hebbian Learning of Temporal Sequences

Author

Tully, Philip J, Lindén, Henrik, Hennig, Matthias H, Lansner, Anders, Tully, Philip J, Lindén, Henrik, Hennig, Matthias H, and Lansner, Anders

Abstract

Many cognitive and motor functions are enabled by the temporal representation and processing of stimuli, but it remains an open issue how neocortical microcircuits can reliably encode and replay such sequences of information. To better understand this, a modular attractor memory network is proposed in which meta-stable sequential attractor transitions are learned through changes to synaptic weights and intrinsic excitabilities via the spike-based Bayesian Confidence Propagation Neural Network (BCPNN) learning rule. We find that the formation of distributed memories, embodied by increased periods of firing in pools of excitatory neurons, together with asymmetrical associations between these distinct network states, can be acquired through plasticity. The model's feasibility is demonstrated using simulations of adaptive exponential integrate-and-fire model neurons (AdEx). We show that the learning and speed of sequence replay depends on a confluence of biophysically relevant parameters including stimulus duration, level of background noise, ratio of synaptic currents, and strengths of short-term depression and adaptation. Moreover, sequence elements are shown to flexibly participate multiple times in the sequence, suggesting that spiking attractor networks of this type can support an efficient combinatorial code. The model provides a principled approach towards understanding how multiple interacting plasticity mechanisms can coordinate hetero-associative learning in unison.

Published
2016

13. Synaptic and nonsynaptic plasticity approximating probabilistic inference

Author

Tully, Philip, Hennig, Matthias, Lansner, Anders, Tully, Philip, Hennig, Matthias, and Lansner, Anders

Abstract

Learning and memory operations in neural circuits are believed to involve molecular cascades of synaptic and nonsynaptic changes that lead to a diverse repertoire of dynamical phenomena at higher levels of processing. Hebbian and homeostatic plasticity, neuromodulation, and intrinsic excitability all conspire to form and maintain memories. But it is still unclear how these seemingly redundant mechanisms could jointly orchestrate learning in a more unified system. To this end, a Hebbian learning rule for spiking neurons inspired by Bayesian statistics is proposed. In this model, synaptic weights and intrinsic currents are adapted on-line upon arrival of single spikes, which initiate a cascade of temporally interacting memory traces that locally estimate probabilities associated with relative neuronal activation levels. Trace dynamics enable synaptic learning to readily demonstrate a spike-timing dependence, stably return to a set-point over long time scales, and remain competitive despite this stability. Beyond unsupervised learning, linking the traces with an external plasticity-modulating signal enables spike-based reinforcement learning. At the postsynaptic neuron, the traces are represented by an activity-dependent ion channel that is shown to regulate the input received by a postsynaptic cell and generate intrinsic graded persistent firing levels. We show how spike-based Hebbian-Bayesian learning can be performed in a simulated inference task using integrate-and-fire (IAF) neurons that are Poisson-firing and background-driven, similar to the preferred regime of cortical neurons. Our results support the view that neurons can represent information in the form of probability distributions, and that probabilistic inference could be a functional by-product of coupled synaptic and nonsynaptic mechanisms operating over several timescales. The model provides a biophysical realization of Bayesian computation by reconciling several observed neural phenomena whose function, This Document is Protected by copyright and was first published by Frontiers. All rights reserved. it is reproduced with permission.QC 20150430, BrainScaleS, Erasmus Mundus EuroSPIN

Published
2014
Full Text
View/download PDF

17. Duleek (B.)

Author

Duleek (B.), Braonáin, Micheál Ó, Gormáin, Séamas Ó, Cionnaith, Tomas Mac, Tulaigh, Pilib Ó, Conchubhair, Seán Ó, Gormáin, Seamas Ó, Cionnaith, Tomás Mac, Gormáin, Seámas Ó, Gormáin, Seumas Ó, Baoighill, Éanna Ó, Cionnaith, Séumas Mac, Cluintiúin, Padraig, Stanáin, Pilib Mac Giolla, Gabhann, Liam Mac, Braonáin, Ronan Ó, Seán, Lochráin, Brian Ó, Pudarlaigh, Peadar, Colmáin, Padraig Ó, Seanan, Pilib Mac Giolla, Moss, Pete, Brennan, Ronan, Gormáin, Séumas Ó, Tulaigh, Pilib Mac An, Kenna, Jane Mc, Tully, Philip, Clarke, Brendan, Cluintiuin, Padraig, Connor, John, Gormáin, Seúmas Ó, Gorman, Seumas, Cionnaith, Seumas Mac, Boyle, Enda O', and Sampson, Seán

Subjects
Halloween, Historic sites, Candlemaking, Folk poetry, local legends, Manners and customs, Winds, Boyne, Battle of the, Ireland, 1690, Animals, Cemeteries, Potatoes, Weather, Geographical myths, Folklore, Treasure troves, Schools, Agriculture, Folk beliefs, Traditional medicine, Supernatural beings, Dwellings, May (Month), Christmas, Death, Ringforts, Land use, Butter, Recreation, Textile industry
Abstract

A collection of folklore and local history stories from Duleek (B.) (school) (Duleek, Co. Meath), collected as part of the Schools' Folklore Scheme, 1937-1938 under the supervision of teacher Micheál Ó Braonáin., Focla Gaeilge atá in Úsáid Fós i mBéal na mBéarlóirí i gCo. na Mí / Braonáin, Micheál Ó -- Ainmneacha Lustraí / Braonáin, Micheál Ó -- Páirceanna / Braonáin, Micheál Ó -- Duleek -- Weather / Gormáin, Séamas Ó -- Weather / Cionnaith, Tomas Mac -- Weather / Tulaigh, Pilib Ó -- Weather / Conchubhair, Seán Ó -- Rocks of Carragubbin, Duleek / Gormáin, Seamas Ó -- Rocks of Carragubbin, Duleek / Conchubhair, Seán Ó -- Ethnographic -- Duleek Ghost Stories / Tulaigh, Pilib Ó -- Duleek Ghost Stories / Cionnaith, Tomás Mac -- Duleek Ghost Stories / Cionnaith, Tomás Mac -- Duleek Ghost Stories / Gormáin, Seamas Ó -- Duleek Ghost Stories / Conchubhair, Seán Ó -- Local Heroes - Collier the Robber / Conchubhair, Seán Ó -- Collier the Robber / Gormáin, Seámas Ó -- Collier the Robber / Tulaigh, Pilib Ó -- Hidden Treasure / Gormáin, Seumas Ó -- Hidden Treasure / Conchubhair, Seán Ó -- Hidden Treasure / Baoighill, Éanna Ó -- Hidden Treasure / Cionnaith, Séumas Mac -- Hidden Treasure / Cluintiúin, Padraig -- Hidden Treasure / Stanáin, Pilib Mac Giolla -- Hidden Treasure / Gabhann, Liam Mac -- Weaving Industry of Duleek of Former Times / Braonáin, Ronan Ó -- Weaving Industry of Duleek of Former Times / Gormáin, Seumas Ó -- Weaving Industry / Conchubhair, Seán Ó -- Old Mills of Duleek / Conchubhair, Seán Ó -- Old Mills of Duleek / Gormáin, Seumas Ó -- Beaumond Mill / , Seán -- Duleek Mill / Lochráin, Brian Ó -- Remains of Old Churches About Duleek / Gormáin, Seumas Ó -- Old Churches of Duleek / Conchubhair, Seán Ó -- Old Churches of Duleek / Braonáin, Ronan Ó -- Remains of Old Churches Around Duleek / Lochráin, Brian Ó -- Remains of Old Churches Around Duleek / Tulaigh, Pilib Ó -- Remains of Old Churches - St Cianán / Pudarlaigh, Peadar -- Remains of Old Churches - St Cianán / Colmáin, Padraig Ó -- St Cianán Patron of Duleek / Gabhann, Liam Mac -- Another Legend of St Cianán / Baoighill, Éanna Ó -- Wells of Duleek - St Cianán / Gormáin, Seumas Ó -- Wells of Duleek- St Cianán's Well / Conchubhair, Seán Ó -- The Wells of Duleek / Braonáin, Ronan Ó -- St Cianán's Well / Lochráin, Brian Ó -- St Cianán's Well / Seanan, Pilib Mac Giolla -- Standing on a Stray Sod / Gormáin, Seumas Ó -- Raspberry Production in Duleek / Moss, Pete -- Raspberry Production in Duleek - Preparation of the Soil / Moss, Pete -- Raspberry Production in Duleek - Planting / Moss, Pete -- Raspberry Production in Duleek - Cultivation / Moss, Pete -- Raspberry Production in Duleek - Manuring / Moss, Pete -- Raspberry Production in Duleek - Harvesting the Crop / Moss, Pete -- Raspberry Production in Duleek - Marketing the Fruit / Moss, Pete -- Cromwell and Duleek / Baoighill, Éanna Ó -- Cromwell and Duleek / Gormáin, Seumas Ó -- Cromwell and Duleek / Conchubhair, Seán Ó -- Cromwell and Duleek / Brennan, Ronan -- Duleek - Games of Former Times / Gormáin, Séumas Ó -- Duleek Games / Tulaigh, Pilib Mac an -- Duleek Games - Jackstones / Kenna, Jane Mc -- May Day Customs in Old Duleek / Gormáin, Seumas Ó -- May Day Customs in Old Duleek / Conchubhair, Seán Ó -- May Day Customs in Old Duleek / Brennan, Ronan -- May Day Customs in Old Duleek / Cionnaith, Tomas Mac -- Remarkable Finds at Keenoge, Duleek / Braonáin, Micheál Ó -- Rushlights / Gormáin, Seumas Ó -- Rushlights / Tully, Philip -- Rushlights / Clarke, Brendan -- Rushlights / Cionnaith, Tomas Mac -- Old Graveyard of Duleek (now disused) / Gormáin, Seumas Ó -- Old Duleek Graveyard - Now Disused / Pudarlaigh, Peadar -- The Caterpillar Known in Duleek as God's Horse / Conchubhair, Seán Ó -- Caterpillar Known in Duleek as God's Horse / Cluintiuin, Padraig -- Duleek and the Battle of the Boyne / Baoighill, Éanna Ó -- Duleek and the Battle of the Boyne / Brennan, Ronan -- Duleek and the Battle of the Boyne / Connor, John -- Wakes and Funerals / Conchubhair, Seán Ó -- Wakes and Funerals / Gormáin, Séumas Ó -- Dally-Iron / Braonáin, Micheál Ó -- Irish Caoin / Braonáin, Micheál Ó -- A list of successive managers of the present national school as far back as memory goes. -- Things Lucky and Unlucky / Gormáin, Séumas Ó -- Things Lucky and Unlucky / Connor, John -- Local Heroes / Gormáin, Séumas Ó -- Hallow Eve Customs in Duleek / Gormáin, Séumas Ó -- Hallow Eve Customs / Brennan, Ronan -- Races on the Commons, Duleek / Gormáin, Séumas Ó -- Potato Crop / Gormáin, Seumas Ó -- Potato Crop / Tulaigh, Pilib Ó -- Big Wind / Gormáin, Seúmas Ó -- Churning / Gorman, Seumas -- Churning / Cionnaith, Seumas Mac -- Christmas / Boyle, Enda O' -- Evictions in Duleek / Gormáin, Seumas Ó -- Evictions in Duleek / Conchubhair, Seán Ó -- Cures / Gormáin, Seumas Ó -- Cures / Brennan, Ronan -- Cures / Conchubhair, Seán Ó -- Cures / Tully, Philip -- Cures / Boyle, Enda O' -- Cures / Cionnaith, Seumas Mac -- Boys' School, Duleek -- Schools, Duleek / Gormáin, Seumas Ó -- Girls' School -- Schools, Duleek / Conchubhair, Seán Ó -- Bronze Age Battle-Axe -- Ruins of Duleek / Braonáin, Micheál Ó -- Weasel / Gormáin, Seumas Ó -- Sally Picker or Pecker - A Bird / Sampson, Seán -- Sally Picker / Cionnaith, Tomás Mac, Supported by funding from the Department of Arts, Heritage and the Gaeltacht (Ireland), University College Dublin, and the National Folklore Foundation (Fondúireacht Bhéaloideas Éireann), 2014-2016.

Published
1937
Full Text
View/download PDF

18. Large-Scale Simulations of Plastic Neural Networks on Neuromorphic Hardware

Author

Knight, James C, Tully, Philip J, Kaplan, Bernhard A, Lansner, Anders, and Furber, Steve B

Subjects
Cellular and Molecular Neuroscience, SpiNNaker, learning, plasticity, fixed-point accuracy, Neuroscience (miscellaneous), event-driven simulation, Anatomy, Bayesian confidence propagation neural network (BCPNN), digital neuromorphic hardware, Neuroscience
Abstract

SpiNNaker is a digital, neuromorphic architecture designed for simulating large-scale spiking neural networks at speeds close to biological real-time. Rather than using bespoke analog or digital hardware, the basic computational unit of a SpiNNaker system is a general-purpose ARM processor, allowing it to be programmed to simulate a wide variety of neuron and synapse models. This flexibility is particularly valuable in the study of biological plasticity phenomena. A recently proposed learning rule based on the Bayesian Confidence Propagation Neural Network (BCPNN) paradigm offers a generic framework for modeling the interaction of different plasticity mechanisms using spiking neurons. However, it can be computationally expensive to simulate large networks with BCPNN learning since it requires multiple state variables for each synapse, each of which needs to be updated every simulation time-step. We discuss the trade-offs in efficiency and accuracy involved in developing an event-based BCPNN implementation for SpiNNaker based on an analytical solution to the BCPNN equations, and detail the steps taken to fit this within the limited computational and memory resources of the SpiNNaker architecture. We demonstrate this learning rule by learning temporal sequences of neural activity within a recurrent attractor network which we simulate at scales of up to 2.0 × 104 neurons and 5.1 × 107 plastic synapses: the largest plastic neural network ever to be simulated on neuromorphic hardware. We also run a comparable simulation on a Cray XC-30 supercomputer system and find that, if it is to match the run-time of our SpiNNaker simulation, the super computer system uses approximately 45× more power. This suggests that cheaper, more power efficient neuromorphic systems are becoming useful discovery tools in the study of plasticity in large-scale brain models.

Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results