Hyper-Dimensional Computing Challenges and Opportunities for AI Applications
Brain-inspired architectures are gaining increased attention, especially for edge devices to perform cognitive tasks utilizing its limited energy budget and computing resources. Hyperdimensional computing (HDC) paradigm is an emerging framework inspired by an abstract representation of neuronal circ...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9354795/ |
| _version_ | 1798031516888989696 |
|---|---|
| author | Eman Hassan Yasmin Halawani Baker Mohammad Hani Saleh |
| author_facet | Eman Hassan Yasmin Halawani Baker Mohammad Hani Saleh |
| author_sort | Eman Hassan |
| collection | DOAJ |
| description | Brain-inspired architectures are gaining increased attention, especially for edge devices to perform cognitive tasks utilizing its limited energy budget and computing resources. Hyperdimensional computing (HDC) paradigm is an emerging framework inspired by an abstract representation of neuronal circuits’ attributes in the human brain. That includes a fully holographic random representation, high-dimension vectors representing data, and robustness to uncertainty. The basic HDC pipeline consists of an encoding, training and comparison stages. The encoding algorithm maps different representations of inputs into a single class and stores them in the associative memory (AM) throughout the training stage. Later, during the inference stage, the similarity is computed between the query vector, which is encoded using the same encoding model, and the stored classes in the AM. HDC has shown promising results for 1D applications using less power, and lower latency than state-of-the-art digital neural networks (DNN). While in 2D applications, convolutional neural network (CNN) still achieves higher classification accuracy at the expense of more computations. In this paper, a comprehensive study on the HDC paradigm, main algorithms, and its implementation is presented. Moreover, the main state-of-the-art HDC architectures for 1D and 2D applications are highlighted. The article also analyzes two competing paradigms, namely, HDC and CNN, in terms of accuracy, complexity, and the number of operations. The paper concluded by highlighting challenges and recommendations for future directions on the HDC framework. |
| first_indexed | 2024-04-11T19:57:41Z |
| format | Article |
| id | doaj.art-0ad7689342ec4d1f9f7db11a72cd735f |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-11T19:57:41Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-0ad7689342ec4d1f9f7db11a72cd735f2022-12-22T04:05:49ZengIEEEIEEE Access2169-35362022-01-0110976519766410.1109/ACCESS.2021.30597629354795Hyper-Dimensional Computing Challenges and Opportunities for AI ApplicationsEman Hassan0https://orcid.org/0000-0002-3750-3336Yasmin Halawani1https://orcid.org/0000-0001-9617-5080Baker Mohammad2https://orcid.org/0000-0002-6063-473XHani Saleh3https://orcid.org/0000-0002-7185-0278Department of Electrical and Computer Engineering, System-on-Chip Center, Khalifa University, Abu-Dhabi, United Arab EmiratesDepartment of Electrical and Computer Engineering, System-on-Chip Center, Khalifa University, Abu-Dhabi, United Arab EmiratesDepartment of Electrical and Computer Engineering, System-on-Chip Center, Khalifa University, Abu-Dhabi, United Arab EmiratesDepartment of Electrical and Computer Engineering, System-on-Chip Center, Khalifa University, Abu-Dhabi, United Arab EmiratesBrain-inspired architectures are gaining increased attention, especially for edge devices to perform cognitive tasks utilizing its limited energy budget and computing resources. Hyperdimensional computing (HDC) paradigm is an emerging framework inspired by an abstract representation of neuronal circuits’ attributes in the human brain. That includes a fully holographic random representation, high-dimension vectors representing data, and robustness to uncertainty. The basic HDC pipeline consists of an encoding, training and comparison stages. The encoding algorithm maps different representations of inputs into a single class and stores them in the associative memory (AM) throughout the training stage. Later, during the inference stage, the similarity is computed between the query vector, which is encoded using the same encoding model, and the stored classes in the AM. HDC has shown promising results for 1D applications using less power, and lower latency than state-of-the-art digital neural networks (DNN). While in 2D applications, convolutional neural network (CNN) still achieves higher classification accuracy at the expense of more computations. In this paper, a comprehensive study on the HDC paradigm, main algorithms, and its implementation is presented. Moreover, the main state-of-the-art HDC architectures for 1D and 2D applications are highlighted. The article also analyzes two competing paradigms, namely, HDC and CNN, in terms of accuracy, complexity, and the number of operations. The paper concluded by highlighting challenges and recommendations for future directions on the HDC framework.https://ieeexplore.ieee.org/document/9354795/Brain-inspired architectureshyperdimensional computingconvolutional neural networksencodingclassificationassociative memory |
| spellingShingle | Eman Hassan Yasmin Halawani Baker Mohammad Hani Saleh Hyper-Dimensional Computing Challenges and Opportunities for AI Applications IEEE Access Brain-inspired architectures hyperdimensional computing convolutional neural networks encoding classification associative memory |
| title | Hyper-Dimensional Computing Challenges and Opportunities for AI Applications |
| title_full | Hyper-Dimensional Computing Challenges and Opportunities for AI Applications |
| title_fullStr | Hyper-Dimensional Computing Challenges and Opportunities for AI Applications |
| title_full_unstemmed | Hyper-Dimensional Computing Challenges and Opportunities for AI Applications |
| title_short | Hyper-Dimensional Computing Challenges and Opportunities for AI Applications |
| title_sort | hyper dimensional computing challenges and opportunities for ai applications |
| topic | Brain-inspired architectures hyperdimensional computing convolutional neural networks encoding classification associative memory |
| url | https://ieeexplore.ieee.org/document/9354795/ |
| work_keys_str_mv | AT emanhassan hyperdimensionalcomputingchallengesandopportunitiesforaiapplications AT yasminhalawani hyperdimensionalcomputingchallengesandopportunitiesforaiapplications AT bakermohammad hyperdimensionalcomputingchallengesandopportunitiesforaiapplications AT hanisaleh hyperdimensionalcomputingchallengesandopportunitiesforaiapplications |