# US 2025/0147725 A1 · Accumulator, Method for Accumulator and Chip Circuit

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0147725 A1
- Document ID: US20250147725A1
- Status: Published application
- Kind: A1
- Application number: 18/650,256
- Filed: 2024-04-30
- Published: 2025-05-08
- Earliest claimed: 2023-11-03
- Assignees: Shenzhen Corerain Technologies Co., Ltd.
- Applicants: Shenzhen Corerain Technologies Co., Ltd.
- Inventors: Chenchen Lu; Kuen Hung Tsoi; Xinyu Niu
- Web page: https://patentgrep.com/patent/US20250147725A1

## CPC

- G06F 7/49915: {Mantissa overflow or underflow in handling floating-point numbers}
- G06F 7/483: Computations with numbers represented by a non-linear combination of denominational numbers, e.g. rational numbers, logarithmic number system or floating-point numbers {(G06F7/4806, G06F7/4824, G06F7/49, G06F7/491, G06F7/544 take precedence)}
- G06F 7/485: Adding; Subtracting {(G06F7/4833, G06F7/4836 take precedence)}
- G06F 7/5443: {Sum of products (for applications thereof, see the relevant places, e.g. G06F17/10, H03H17/00)}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

The present application provides an accumulator, a method for the accumulator and a chip circuit. The accumulator includes an accumulation execution module, a difference correction module and a first register, wherein: the accumulation execution module is used to perform accumulation operations on input data of floating point type, and the truncated data in each operation is reserved as truncation error, and the truncation error is fed back to the difference correction module; the difference correction module uses the fed-back truncation error to perform superposition calculation on the external input data, and outputs the superimposed input data to the accumulation execution module; the first register is used to buffer the accumulation result of the accumulation execution module, and the accumulation result is sent back to the accumulation execution module through a feedback channel. It is possible to improve the calculation efficiency while improving the accumulation accuracy of floating point numbers.

## Claims

### Claim 1

An accumulator for floating-point number accumulation, characterized in that
the accumulator comprises an accumulation execution module, a difference correction module and a first register, wherein:
the accumulation execution module is used to perform an accumulation operation on a floating-point type input data, and a truncated data in each operation is reserved as a truncation error, and the truncation error is fed back to the difference correction module;
the difference correction module uses the fed-back truncation error to perform a superposition calculation on an external input data, and outputs a superimposed input data to the accumulation execution module; and
the first register is used to buffer an accumulation result of the accumulation execution module, and the accumulation result is sent back to the accumulation execution module through a feedback channel.

### Claim 2 (depends on 1)

The accumulator according to claim 1, characterized by
further comprising: a second register for buffering the truncation error and providing the truncation error to the difference correction module.

### Claim 3 (depends on 1)

The accumulator according to claim 1, characterized by
further comprising: a third register for buffering the superimposed input data and providing the superimposed input data to the accumulation execution module.

### Claim 4 (depends on 2)

The accumulator according to claim 2, characterized in that
the accumulation execution module is configured to align the accumulation result with the superimposed input data;
store a truncated mantissa during the alignment as the truncation error; and
perform an alignment addition on the truncated floating-point data, and buffer a result as the accumulation result into the first register.

### Claim 5 (depends on 4)

The accumulator according to claim 4, characterized in that
the accumulation execution module comprises:
an adder used to perform an addition operation on the accumulation result and the superimposed input data; and
a truncation error register used to store the truncation error calculation result.

### Claim 6 (depends on 5)

The accumulator according to claim 5, characterized in that
the accumulation execution module is configured to
in response to the accumulation result exceeding a number of alignment bits, perform a normalized truncation on the accumulation result, and truncate a high-bit data of corresponding alignment bits as the accumulation result and store the accumulation result in the first register;
concatenate a low bit truncated after the normalized truncation with a mantissa truncated by an alignment truncation, and store a result data obtained by the concatenation in the second register as the truncation error.

### Claim 7

A method for an accumulator, characterized in that
the method comprises:
calculating an accumulation result of a floating point data and obtaining a truncation error;
feeding the truncation error back to an input end; and
using the fed-back truncation error to perform a superposition calculation on an external input data, and outputting the superimposed input data for accumulation calculation.

### Claim 8 (depends on 7)

The method according to claim 7, characterized in that
the step of calculating an accumulation result of a floating point data and obtaining a truncation error comprises:
during the accumulation, performing an alignment addition on the accumulation result and the superimposed input data, using a calculation result as the accumulation result, and using a mantissa truncated by the alignment as the truncation error;
in response to there being a carry in the accumulation result, performing a normalized truncation on the data, and concatenating a low bit truncated after the normalized truncation with the mantissa truncated by the alignment truncation to form the truncation error; and
only after the accumulation is completed, performing a rounding operation on the accumulation result.

### Claim 9 (depends on 1)

A chip circuit, characterized by comprising:
the accumulator according to claim 1.

### Claim 10 (depends on 2)

A chip circuit, characterized by comprising:
the accumulator according to claim 2.

### Claim 11 (depends on 3)

A chip circuit, characterized by comprising:
the accumulator according to claim 3.

### Claim 12 (depends on 4)

A chip circuit, characterized by comprising:
the accumulator according to claim 4.

### Claim 13 (depends on 5)

A chip circuit, characterized by comprising:
the accumulator according to claim 5.

### Claim 14 (depends on 6)

A chip circuit, characterized by comprising:
the accumulator according to claim 6.
