Sökning: id:"swepub:oai:DiVA.org:su-117399" >
Reducing the comput...
-
Vogginger, BernhardTechnical University Dresden
(författare)
Reducing the computational footprint for real-time BCPNN learning
- Artikel/kapitelEngelska2015
Förlag, utgivningsår, omfång ...
-
2015-01-22
-
Frontiers Media SA,2015
-
printrdacarrier
Nummerbeteckningar
-
LIBRIS-ID:oai:DiVA.org:su-117399
-
https://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-117399URI
-
https://doi.org/10.3389/fnins.2015.00002DOI
-
https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-165947URI
Kompletterande språkuppgifter
-
Språk:engelska
-
Sammanfattning på:engelska
Ingår i deldatabas
Klassifikation
-
Ämneskategori:ref swepub-contenttype
-
Ämneskategori:art swepub-publicationtype
Anmärkningar
-
AuthorCount:6;
-
QC 20150506
-
The implementation of synaptic plasticity in neural simulation or neuromorphic hardware is usually very resource-intensive, often requiring a compromise between efficiency and flexibility. A versatile, but computationally-expensive plasticity mechanism is provided by the Bayesian Confidence Propagation Neural Network (BCPNN) paradigm. Building upon Bayesian statistics, and having clear links to biological plasticity processes, the BCPNN learning rule has been applied in many fields, ranging from data classification, associative memory, reward-based learning, probabilistic inference to cortical attractor memory networks. In the spike-based version of this learning rule the pre-, postsynaptic and coincident activity is traced in three low-pass-filtering stages, requiring a total of eight state variables, whose dynamics are typically simulated with the fixed step size Euler method. We derive analytic solutions allowing an efficient event-driven implementation of this learning rule. Further speedup is achieved by first rewriting the model which reduces the number of basic arithmetic operations per update to one half, and second by using look-up tables for the frequently calculated exponential decay. Ultimately, in a typical use case, the simulation using our approach is more than one order of magnitude faster than with the fixed step size Euler method. Aiming for a small memory footprint per BCPNN synapse, we also evaluate the use of fixed-point numbers for the state variables, and assess the number of bits required to achieve same or better accuracy than with the conventional explicit Euler method. All of this will allow a real-time simulation of a reduced cortex model based on BCPNN in high performance computing. More important, with the analytic solution at hand and due to the reduced memory bandwidth, the learning rule can be efficiently implemented in dedicated or existing digital neuromorphic hardware.
Ämnesord och genrebeteckningar
Biuppslag (personer, institutioner, konferenser, titlar ...)
-
Schueffny, ReneTechnical University Dresden
(författare)
-
Lansner, AndersKTH,Stockholms universitet,Numerisk analys och datalogi (NADA),Royal Institute of Technology (KTH), Sweden,Beräkningsbiologi, CB,Stockholm University, Sweden,Lansner(Swepub:kth)u12s8cr8
(författare)
-
Cederström, LoveTechnical University Dresden
(författare)
-
Partzsch, JohannesTechnical University Dresden
(författare)
-
Hoeppner, SebastianTechnical University Dresden
(författare)
-
Technical University DresdenNumerisk analys och datalogi (NADA)
(creator_code:org_t)
Sammanhörande titlar
-
Ingår i:Frontiers in Neuroscience: Frontiers Media SA91662-45481662-453X1662-6443
Internetlänk
Hitta via bibliotek
Till lärosätets databas