摘要:
A method for context-aware text recognition employing two neuromorphic computing models, auto-associative neural network and cogent confabulation. The neural network model performs the character recognition from input image and produces one or more candidates for each character in the text image input. The confabulation models perform the context-aware text extraction and completion, based on the character recognition outputs and the word and sentence knowledge bases.
摘要:
An apparatus for a hybrid architecture that consists of a general purpose microprocessor and a hardware accelerator for accelerating the discovery of DNA reverse complement, edit distance codes. Two embodiments are implemented and evaluated, including a code generator that uses a genetic algorithm (GA) to produce nearly locally optimal codes in a few minutes, and a code extender that uses exhaustive search to produce locally optimum codes in about 1.5 hours for the case of length 16 codes. Experimental results demonstrate that the GA embodiment can find ˜99% of the words in locally optimum libraries, and that the hybrid architecture embodiment provides more than 1000 times speed-up compared to a software only implementation.
摘要:
An apparatus for a hybrid architecture that consists of a general purpose microprocessor and a hardware accelerator for accelerating the discovery of DNA reverse complement, edit distance codes. Two embodiments are implemented and evaluated, including a code generator that uses a genetic algorithm (GA) to produce nearly locally optimal codes in a few minutes, and a code extender that uses exhaustive search to produce locally optimum codes in about 1.5 hours for the case of length 16 codes. Experimental results demonstrate that the GA embodiment can find ˜99% of the words in locally optimum libraries, and that the hybrid architecture embodiment provides more than 1000 times speed-up compared to a software only implementation.
摘要:
An apparatus that accelerates the determination of NN free energy of binding estimates for a large number of DNA oligomers using reconfigurable hardware and applies it to the design of high quality DNA code word libraries. The invention provides a reconfigurable hardware accelerator and method for implementing a nearest-neighbor based free energy calculation. The invention further provides a method to produce the maximum weight of the 2-stem common subsequence of two DNA oligonucleotides. In practice, the present invention comprises a general purpose microprocessor or computer, a hardware accelerator, and a software program.
摘要:
An apparatus for a hybrid architecture that consists of a general purpose microprocessor and a hardware accelerator for accelerating the discovery of DNA reverse complement edit distance codes. An embodiment is implemented and evaluated, including a code extender that uses exhaustive search to produce locally optimum codes in about 1.5 hours for the case of length 16 codes.
摘要:
An apparatus for a hybrid architecture that consists of a general purpose microprocessor and a hardware accelerator for accelerating the discovery of DNA reverse complement, edit distance codes. Two embodiments are implemented and evaluated, including a code generator that uses a genetic algorithm (GA) to produce nearly locally optimal codes in a few minutes, and a code extender that uses exhaustive search to produce locally optimum codes in about 1.5 hours for the case of length 16 codes. Experimental results demonstrate that the GA embodiment can find ˜99% of the words in locally optimum libraries, and that the hybrid architecture embodiment provides more than 1000 times speed-up compared to a software only implementation.
摘要:
An apparatus for a hybrid architecture that consists of a general purpose microprocessor and a hardware accelerator for accelerating the discovery of DNA reverse complement, edit distance codes. Two embodiments are implemented and evaluated, including a code generator that uses a genetic algorithm (GA) to produce nearly locally optimal codes in a few minutes, and a code extender that uses exhaustive search to produce locally optimum codes in about 1.5 hours for the case of length 16 codes. Experimental results demonstrate that the GA embodiment can find ˜99% of the words in locally optimum libraries, and that the hybrid architecture embodiment provides more than 1000 times speed-up compared to a software only implementation.