ALAMO: FPGA acceleration of deep learning algorithms with a modularized RTL compiler.
Yufei MaNaveen SudaYu CaoSarma B. K. VrudhulaJae-sun SeoPublished in: Integr. (2018)
Keyphrases
- learning algorithm
- deep architectures
- deep learning
- real time image processing
- training data
- programming language
- high speed
- machine learning algorithms
- learning problems
- active learning
- hardware design
- hardware description language
- machine learning
- field programmable gate array
- hardware implementation
- general purpose
- verilog hdl
- learning process
- learning scheme
- unsupervised learning
- rule base
- learning tasks
- fpga technology
- learning rate
- software systems
- real time
- level parallelism
- fpga implementation
- data acquisition
- single chip
- low cost
- signal processing
- efficient learning
- membership queries
- hardware architecture
- parallel programming
- highly optimized
- reinforcement learning
- digital signal
- distributed memory machines
- back propagation