Warning: pg_query(): Query failed: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 in /dati/webiit-old/includes/database.pgsql.inc on line 138 Warning: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 query: SELECT data, created, headers, expire, serialized FROM cache_page WHERE cid = 'https://www-old.iit.cnr.it/node/59639' in /dati/webiit-old/includes/database.pgsql.inc on line 159 Warning: pg_query(): Query failed: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 in /dati/webiit-old/includes/database.pgsql.inc on line 138 Warning: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 query: SELECT data, created, headers, expire, serialized FROM cache_page WHERE cid = 'https://www-old.iit.cnr.it/node/59639' in /dati/webiit-old/includes/database.pgsql.inc on line 159 Dynamic Hard Pruning of Neural Networks at the Edge of the Internet | IIT - CNR - Istituto di Informatica e Telematica
IIT Home Page CNR Home Page

Dynamic Hard Pruning of Neural Networks at the Edge of the Internet

Neural Networks (NN), although successfully applied to several Artificial Intelligence tasks, are often unnecessarily over-parametrized. In fog/edge computing, this might make their training prohibitive on resourceconstrained devices, contrasting with the current trend of decentralising intelligence from remote data-centres to local constrained devices. Therefore, we investigate the problem of training effective NN models on constrained devices having a fixed, potentially small, memory budget. We target techniques that are both resource-efficient and performance effective, while enabling significant network compression. Our technique, called Dynamic Hard Pruning (DynHP), incrementally prunes the network during training, identifying neurons that marginally contribute to the model accuracy. DynHP enables a tunable size reduction of the final neural network and reduces the NN memory occupancy during training. Freed memory is reused by a dynamic batch sizing approach to counterbalance the accuracy degradation caused by the hard pruning strategy, improving its convergence and effectiveness. We assess the performance of DynHP through reproducible experiments on two public datasets, comparing them against reference competitors. Results show that DynHP compresses a NN up to 10 times without significant performance drops (up to 5% relative error w.r.t. competitors), reducing up to 80% the training memory occupancy


2020

Autori esterni: Franco Maria Nardini, Raffaele Perego (ISTI-CNR), Raffaele Perego (ISTI-CNR)
Autori IIT:

Tipo: Rapporto Tecnico
Area di disciplina: Computer Science & Engineering
IIT TR-21/2020

File: IIT-21-2020.pdf

Attività: Future Internet