dc.contributor |
Graduate Program in Computer Engineering. |
|
dc.contributor.advisor |
Yurdakul, Arda. |
|
dc.contributor.author |
Bilgili, Emre. |
|
dc.date.accessioned |
2023-10-15T06:58:19Z |
|
dc.date.available |
2023-10-15T06:58:19Z |
|
dc.date.issued |
2022 |
|
dc.identifier.other |
CMPE 2022 B55 |
|
dc.identifier.uri |
http://digitalarchive.boun.edu.tr/handle/123456789/19721 |
|
dc.description.abstract |
The execution time, resource and energy costs of deep learning applications become much more important as their popularity grows. The Constant Matrix Multi plication has been studied for a long time and takes place in deep learning applications. Reducing the computation cost of those applications is a highly active research topic. The weights are pruned or quantized while satisfying the desired accuracy requirement. The pruned matrices are compressed into one-dimensional arrays without data loss. Matrix multiplication is performed by processing those arrays without decompression. Processing one-dimensional arrays to perform matrix multiplication is deployed on vari ous hardware platforms that employ Central Processing Unit, Graphics Processor Unit and Field-Programmable Gate Array. The deployments can also be supported with common subexpression elimination methods to reduce the number of multiplications, additions and storage size. However, the state-of-the-art methods do not scale well for the large constant matrices as they reach hours for extracting common subexpressions in a 200 × 200 matrix. In this thesis, a random search-based common subexpression elimination method is constructed to reduce the run-time of the algorithm. The algo rithm produces an adder tree for a 1000 × 1000 matrix in a minute. The Compressed Sparse Row format is extended to build a one-dimensional compression notation for the proposed method. Simulations for a single-core embedded system show that the latency is reduced by 80% for a given 100×100 matrix compared to the state-of-the- art methods. The storage size of the sparse matrices is also reduced by more than half in the experiments compared to the Compressed Sparse Row format. |
|
dc.publisher |
Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2022. |
|
dc.subject.lcsh |
Matrices. |
|
dc.subject.lcsh |
Deep learning (Machine learning) |
|
dc.title |
A common subexpression elimination-based compression method for the constant matrix multipication |
|
dc.format.pages |
xi, 58 leaves |
|