18245843. SCALAR PRODUCT CIRCUIT, AND METHOD FOR COMPUTING BINARY SCALAR PRODUCTS OF AN INPUT VECTOR AND WEIGHT VECTORS simplified abstract (Robert Bosch GmbH)
SCALAR PRODUCT CIRCUIT, AND METHOD FOR COMPUTING BINARY SCALAR PRODUCTS OF AN INPUT VECTOR AND WEIGHT VECTORS
Organization Name
Inventor(s)
Andre Guntoro of Weil Der Stadt (DE)
Taha Ibrahim Ibrahim Soliman of Renningen (DE)
Tobias Kirchner of Ludwigsburg (DE)
SCALAR PRODUCT CIRCUIT, AND METHOD FOR COMPUTING BINARY SCALAR PRODUCTS OF AN INPUT VECTOR AND WEIGHT VECTORS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18245843 titled 'SCALAR PRODUCT CIRCUIT, AND METHOD FOR COMPUTING BINARY SCALAR PRODUCTS OF AN INPUT VECTOR AND WEIGHT VECTORS
Simplified Explanation
The abstract describes a patent application for a scalar product circuit that computes a binary scalar product of an input vector and a weight vector. The circuit includes one or more adders and at least one matrix circuit with memory cells arranged in a matrix format. Each memory cell has a first and second memory state. The matrix circuit also includes weight ranges with multiple bit sections, each connected to an analog-to-digital converter and a bit shifting unit. Each bit section's column lines are connected to the analog-to-digital converter, and a column selection switching element is provided for each column. The bit shifting units are connected to one of the adders, with those in the same weight range connected to the same adder.
- The patent application describes a circuit for computing a binary scalar product of input and weight vectors.
- The circuit includes adders and a matrix circuit with memory cells arranged in a matrix format.
- Each memory cell has two memory states.
- The matrix circuit includes weight ranges with multiple bit sections.
- Each bit section is connected to an analog-to-digital converter and a bit shifting unit.
- The column lines of each bit section are connected to the analog-to-digital converter.
- Each column has a column selection switching element.
- The bit shifting units are connected to one of the adders.
- Bit shifting units in the same weight range are connected to the same adder.
Potential applications of this technology:
- Machine learning and artificial intelligence algorithms that require binary scalar product computations.
- Signal processing applications that involve binary scalar product calculations.
- Cryptography and encryption algorithms that utilize binary scalar products.
Problems solved by this technology:
- Efficient computation of binary scalar products.
- Reduction of computational complexity in scalar product calculations.
- Improved performance and speed in applications that rely on scalar product computations.
Benefits of this technology:
- Faster and more efficient computation of binary scalar products.
- Reduced power consumption due to optimized circuit design.
- Improved accuracy and precision in scalar product calculations.
- Scalability for handling large input and weight vectors.
Original Abstract Submitted
A scalar product circuit for computing a binary scalar product of an input vector and a weight vector. The scalar product circuit includes one or multiple adders and at least one matrix circuit including memory cells that are arranged in multiple rows and multiple columns in the form of a matrix, each memory cell including a first memory state and a second memory state. Each matrix circuit includes at least one weight range including one or multiple bit sections, the matrix circuit including an analog-to-digital converter and a bit shifting unit connected thereto for each bit section, the column lines of the bit section being connected to the analog-to-digital converter, and a column selection switching element being provided for each column. The bit shifting units are connected to one of the adders, those bit shifting units that are included in a weight range being connected to the same adder.
