Recurrent Neural Networks With Column-Wise Matrix-Vector Multiplication on FPGAs.
Zhiqiang QueHiroki NakaharaEriko NurvitadhiAndrew BoutrosHongxiang FanChenglong ZengJiuxi MengKuen Hung TsoiXinyu NiuWayne LukPublished in: IEEE Trans. Very Large Scale Integr. Syst. (2022)
Keyphrases
- recurrent neural networks
- rows and columns
- sparse matrix
- feed forward
- neural network
- complex valued
- matrix multiplication
- data matrix
- reservoir computing
- eigenvalues and eigenvectors
- echo state networks
- feedforward neural networks
- matrix representation
- binary matrix
- recurrent networks
- long short term memory
- floating point
- neural model
- transformation matrix
- binary matrices
- artificial neural networks
- covariance matrix
- nonlinear dynamic systems
- cascade correlation
- linear algebra
- low rank
- matrix factorization
- vector space
- decision making
- artificial intelligence
- symmetric matrices
- long term
- signal processing
- singular value decomposition