Patents by Inventor Tarek M. Taha
Tarek M. Taha has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 10671914Abstract: An analog neuromorphic circuit is disclosed, having input voltages applied to a plurality of inputs of the analog neuromorphic circuit. The circuit also includes a plurality of resistive memories that provide a resistance to each input voltage applied to each of the inputs so that each input voltage is multiplied in parallel by the corresponding resistance of each corresponding resistive memory to generate a corresponding current for each input voltage and each corresponding current is added in parallel. The circuit also includes at least one output signal that is generated from each of the input voltages multiplied in parallel with each of the corresponding currents for each of the input voltages added in parallel. The multiplying of each input voltage with each corresponding resistance is executed simultaneously with adding each corresponding current for each input voltage.Type: GrantFiled: November 11, 2019Date of Patent: June 2, 2020Assignee: University of DaytonInventors: Chris Yakopcic, Md Raqibul Hasan, Tarek M. Taha
-
Patent number: 10622064Abstract: A crossbar circuit determines a match of N bits of data to at least one of M target words simultaneously. The circuit comprises N inputs (one per data bit) and M outputs (one per target word). For each of the M target words, the circuit comprises N?1 biased bits, where each biased bit includes a first data memristor coupled to a corresponding one of the N inputs; a second data memristor coupled to the corresponding one of the N inputs, where the corresponding one of the N inputs is inverted before reaching the second data memristor; and two biasing memristors. Further, the circuit comprises a general bit comprising a first data memristor coupled to the input that does not correspond to any of the biased bits; and a second data memristor coupled to the input that does not correspond to any of the biased bits and is inverted.Type: GrantFiled: January 25, 2019Date of Patent: April 14, 2020Assignee: University of DaytonInventors: Tarek M. Taha, Chris Yakopcic
-
Publication number: 20200074291Abstract: An analog neuromorphic circuit is disclosed, having input voltages applied to a plurality of inputs of the analog neuromorphic circuit. The circuit also includes a plurality of resistive memories that provide a resistance to each input voltage applied to each of the inputs so that each input voltage is multiplied in parallel by the corresponding resistance of each corresponding resistive memory to generate a corresponding current for each input voltage and each corresponding current is added in parallel. The circuit also includes at least one output signal that is generated from each of the input voltages multiplied in parallel with each of the corresponding currents for each of the input voltages added in parallel. The multiplying of each input voltage with each corresponding resistance is executed simultaneously with adding each corresponding current for each input voltage.Type: ApplicationFiled: November 11, 2019Publication date: March 5, 2020Inventors: Chris Yakopcic, Md Raqibul Hasan, Tarek M. Taha
-
Patent number: 10474948Abstract: An analog neuromorphic circuit is disclosed, having input voltages applied to a plurality of inputs of the analog neuromorphic circuit. The circuit also includes a plurality of resistive memories that provide a resistance to each input voltage applied to each of the inputs so that each input voltage is multiplied in parallel by the corresponding resistance of each corresponding resistive memory to generate a corresponding current for each input voltage and each corresponding current is added in parallel. The circuit also includes at least one output signal that is generated from each of the input voltages multiplied in parallel with each of the corresponding currents for each of the input voltages added in parallel. The multiplying of each input voltage with each corresponding resistance is executed simultaneously with adding each corresponding current for each input voltage.Type: GrantFiled: March 28, 2016Date of Patent: November 12, 2019Assignee: University of DaytonInventors: Chris Yakopcic, Raqibul Hasan, Tarek M. Taha
-
Publication number: 20190332930Abstract: An analog neuromorphic circuit is disclosed having resistive memories that provide a resistance to each corresponding input voltage signal. Input voltages are applied to the analog neuromorphic circuit. Each input voltage represents a vector value that is a non-binary value included in a vector that is incorporated into a dot-product operation with weighted matrix values included in a weighted matrix. A controller pairs each resistive memory with another resistive memory. The controller converts each pair of resistance values to a single non-binary value. Each single non-binary value is mapped to a weighted matrix value included in the weighted matrix that is incorporated into the dot-product operation with the vector values included in the vector. The controller generates dot-product operation values from the dot-product operation with the vector and the weighted matrix where each dot-product operation is a non-binary value.Type: ApplicationFiled: July 9, 2019Publication date: October 31, 2019Inventors: Chris Yakopcic, Tarek M. Taha, Md Raqibul Hasan
-
Publication number: 20190228822Abstract: A crossbar circuit determines a match of N bits of data to at least one of M target words simultaneously. The circuit comprises N inputs (one per data bit) and M outputs (one per target word). For each of the M target words, the circuit comprises N?1 biased bits, where each biased bit includes a first data memristor coupled to a corresponding one of the N inputs; a second data memristor coupled to the corresponding one of the N inputs, where the corresponding one of the N inputs is inverted before reaching the second data memristor; and two biasing memristors. Further, the circuit comprises a general bit comprising a first data memristor coupled to the input that does not correspond to any of the biased bits; and a second data memristor coupled to the input that does not correspond to any of the biased bits and is inverted.Type: ApplicationFiled: January 25, 2019Publication date: July 25, 2019Inventors: Tarek M. Taha, Chris Yakopcic
-
Patent number: 10346738Abstract: An analog neuromorphic circuit is disclosed having resistive memories that provide a resistance to each corresponding input voltage signal. Input voltages are applied to the analog neuromorphic circuit. Each input voltage represents a vector value that is a non-binary value included in a vector that is incorporated into a dot-product operation with weighted matrix values included in a weighted matrix. A controller pairs each resistive memory with another resistive memory. The controller converts each pair of resistance values to a single non-binary value. Each single non-binary value is mapped to a weighted matrix value included in the weighted matrix that is incorporated into the dot-product operation with the vector values included in the vector. The controller generates dot-product operation values from the dot-product operation with the vector and the weighted matrix where each dot-product operation is a non-binary value.Type: GrantFiled: January 8, 2019Date of Patent: July 9, 2019Assignee: University of DaytonInventors: Chris Yakopcic, Tarek M. Taha, Md Raqibul Hasan
-
Publication number: 20190138894Abstract: An analog neuromorphic circuit is disclosed having resistive memories that provide a resistance to each corresponding input voltage signal. Input voltages are applied to the analog neuromorphic circuit. Each input voltage represents a vector value that is a non-binary value included in a vector that is incorporated into a dot-product operation with weighted matrix values included in a weighted matrix. A controller pairs each resistive memory with another resistive memory. The controller converts each pair of resistance values to a single non-binary value. Each single non-binary value is mapped to a weighted matrix value included in the weighted matrix that is incorporated into the dot-product operation with the vector values included in the vector. The controller generates dot-product operation values from the dot-product operation with the vector and the weighted matrix where each dot-product operation is a non-binary value.Type: ApplicationFiled: January 8, 2019Publication date: May 9, 2019Inventors: Chris Yakopcic, Tarek M. Taha, Md Raqibul Hasan
-
Patent number: 10176425Abstract: An analog neuromorphic circuit is disclosed having resistive memories that provide a resistance to each corresponding input voltage signal. Input voltages are applied to the analog neuromorphic circuit. Each input voltage represents a vector value that is a non-binary value included in a vector that is incorporated into a dot-product operation with weighted matrix values included in a weighted matrix. A controller pairs each resistive memory with another resistive memory. The controller converts each pair of resistance values to a single non-binary value. Each single non-binary value is mapped to a weighted matrix value included in the weighted matrix that is incorporated into the dot-product operation with the vector values included in the vector. The controller generates dot-product operation values from the dot-product operation with the vector and the weighted matrix where each dot-product operation is a non-binary value.Type: GrantFiled: July 14, 2017Date of Patent: January 8, 2019Assignee: University of DaytonInventors: Chris Yakopcic, Tarek M. Taha, Md Raqibul Hasan
-
Publication number: 20180018559Abstract: An analog neuromorphic circuit is disclosed having resistive memories that provide a resistance to each corresponding input voltage signal. Input voltages are applied to the analog neuromorphic circuit. Each input voltage represents a vector value that is a non-binary value included in a vector that is incorporated into a dot-product operation with weighted matrix values included in a weighted matrix. A controller pairs each resistive memory with another resistive memory. The controller converts each pair of resistance values to a single non-binary value. Each single non-binary value is mapped to a weighted matrix value included in the weighted matrix that is incorporated into the dot-product operation with the vector values included in the vector. The controller generates dot-product operation values from the dot-product operation with the vector and the weighted matrix where each dot-product operation is a non-binary value.Type: ApplicationFiled: July 14, 2017Publication date: January 18, 2018Inventors: Chris Yakopcic, Tarek M. Taha, Md Raqibul Hasan
-
Publication number: 20170011290Abstract: An analog neuromorphic circuit is disclosed having resistive memories that provide a resistance to an input voltage signal as the input voltage signal propagates through the resistive memories generating a first output voltage signal and to provide a resistance to a first error signal that propagates through the resistive memories generating a second output voltage signal. A comparator generates the first error signal that is representative of a difference between the first output voltage signal and the desired output signal and generates the first error signal so that the first error signal propagates back through the plurality of resistive memories. A resistance adjuster adjusts a resistance value associated with each resistive memory based on the first error signal and the second output voltage signal to decrease the difference between the first output voltage signal and the desired output signal.Type: ApplicationFiled: July 6, 2016Publication date: January 12, 2017Inventors: Tarek M. Taha, Raqibul Hasan, Chris Yakopcic
-
Publication number: 20160284400Abstract: An analog neuromorphic circuit is disclosed, having input voltages applied to a plurality of inputs of the analog neuromorphic circuit. The circuit also includes a plurality of resistive memories that provide a resistance to each input voltage applied to each of the inputs so that each input voltage is multiplied in parallel by the corresponding resistance of each corresponding resistive memory to generate a corresponding current for each input voltage and each corresponding current is added in parallel. The circuit also includes at least one output signal that is generated from each of the input voltages multiplied in parallel with each of the corresponding currents for each of the input voltages added in parallel. The multiplying of each input voltage with each corresponding resistance is executed simultaneously with adding each corresponding current for each input voltage.Type: ApplicationFiled: March 28, 2016Publication date: September 29, 2016Inventors: Chris Yakopcic, Raqibul Hasan, Tarek M. Taha
-
Publication number: 20150213884Abstract: A resistive memory array partitioned into a plurality of memory units is disclosed. Each memory unit includes a plurality of resistive memory elements, a plurality of row lines, a plurality of column lines, a plurality of row select switching devices, and a plurality of column select switching devices. Each resistive memory element is in communication with one of the row lines and one of the column lines. Each row line is in communication with a corresponding one of the row select switching devices. Each column line is in communication with a corresponding one of the column select switching devices.Type: ApplicationFiled: January 30, 2014Publication date: July 30, 2015Applicant: UNIVERSITY OF DAYTONInventors: Tarek M. Taha, Chris Yakopcic