SynergicLearning: Neural Network-Based Feature Extraction for Highly-Accurate Hyperdimensional Learning

Machine learning models differ in terms of accuracy, computational/memory complexity, training time, and adaptability among other characteristics. For example, neural networks (NNs) are well-known for their high accuracy due to the quality of their automatic feature extraction while brain-inspired h...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Digest of technical papers - IEEE/ACM International Conference on Computer-Aided Design s. 1 - 9
Hlavní autoři:	Nazemi, Mahdi, Esmaili, Amirhossein, Fayyazi, Arash, Pedram, Massoud
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	Association on Computer Machinery 02.11.2020
Témata:	adaptability Adaptation models arbitrary NN Artificial neural networks automatic feature extraction brain-inspired hyperdimensional learning models classification components computational efficiency Computational modeling Data models end-to-end hardware realization feature extraction feature extractor HD computing framework HD learning models high-performance HD highly-accurate hyperdimensional learning hybrid machine learning model hybrid model Hyperdimensional Learning learning (artificial intelligence) machine learning models neural nets neural network-based feature extraction neural networks On-line Learning online learning Parallel Processing parameterized hardware implementation synergic machine learning model synergic models Training Training data training time
ISSN:	1558-2434
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Machine learning models differ in terms of accuracy, computational/memory complexity, training time, and adaptability among other characteristics. For example, neural networks (NNs) are well-known for their high accuracy due to the quality of their automatic feature extraction while brain-inspired hyperdimensional (HD) learning models are famous for their quick training, computational efficiency, and adaptability. This work presents a hybrid, synergic machine learning model that excels at all the said characteristics and is suitable for incremental, on-line learning on a chip. The proposed model comprises an NN and a classifier. The NN acts as a feature extractor and is specifically trained to work well with the classifier that employs the HD computing framework. This work also presents a parameterized hardware implementation of the said feature extraction and classification components while introducing a compiler that maps any arbitrary NN and/or classifier to the aforementioned hardware. The proposed hybrid machine learning model has the same level of accuracy (i.e. ±1%) as NNs while achieving at least 10% improvement in accuracy compared to HD learning models. Additionally, the end-to-end hardware realization of the hybrid model improves power efficiency by 1.60x compared to state-of-the-art, high-performance HD learning implementations while improving latency by 2.13x. These results have profound implications for the application of such synergic models in challenging cognitive tasks.
ISSN:	1558-2434
DOI:	10.1145/3400302.3415696