Sparse matrix vector multiplication calculation unit for arranged block diagonal weight matrix

Abstract

The invention relates to a sparse matrix vector multiplication calculation unit for an arranged block diagonal weight matrix. The sparse matrix vector multiplication calculation unit comprises a plurality of processing units and an accumulator, and the output of the processing unit is connected with the accumulator. According to the sparse matrix vector multiplication calculation unit for the arranged block diagonal weight matrix provided by the invention, the sparse characteristic of the weight matrix after pruning is fully utilized, and multiplication operation between a zero-value weight and a corresponding input excitation element is avoided. Zero hopping operations can be dynamically enabled in conjunction with sparsity of input excitations. The sparsity of the intermediate product obtained by multiplying the weight by the corresponding input excitation is fully utilized, and the accumulation operation between the zero product and the corresponding product is avoided. The designedpointer generator eliminates the storage overhead of pointers recording non-zero value position information.

Description

technical field [0001] The invention relates to a sparse matrix-vector multiplication calculation unit for arranged block diagonal weight matrices, which belongs to the technical field of integrated circuit design. Background technique [0002] In recent years, deep learning algorithms have begun to replace traditional algorithms and become mainstream algorithms in many fields due to their excellent performance. However, the current mainstream processors (CPU, GPU, DSP, etc.) are still unable to better adapt to their data-intensive computing characteristics. Therefore, there has been a wave of research on deep learning processors in the academic and industrial circles. [0003] The fully connected operation is a very important type of operation in the neural network algorithm, and it is also a type of operation with a very large amount of calculation. Therefore, efficiently performing such operations becomes the key to improving the performance of deep learning processors. ...

AI Technical Summary

Problems solved by technology

On the contrary, because of the addition of additional comparison operations, it will also increase the power consumption of the entire system

(3) The calculation unit does not consider the sparsity of the intermediate product, because the product generated by zero-value weight or zero-value excitation is also zero, so the intermediate product has the same or greater sparsity than the weight matrix

Images