Taiwan Semiconductor Manufacturing Comp. Ltd.

Tsinghua University

Ite Tech. Inc.

Food Industry Research and Development Institute

Ta-Wei Liu

Der-Chyang Yeh

Po-Hao Tsai

An-Jhih Su

Chi-Hsi Wu

Meng-Fan Chang

Li-Hsien Huang

Ming-Shih Yeh

Wei-Yu Chen

Yi-Chung Chou

Fu-Chun Chang

Yen-Lin Chen

Gwo-Fang Yuan

Yen-Hsiang Huang

Cheng-Xin Xue

Yen-Cheng Chiu

Wei-yu Chen

Hsu-Min Chen

Kai-Ping Chen

Hui-Ping Chen

Chih-Yuan Kuo

Ming-Jen Cheng

Ming-Der Wu

Yu-Shan Lin

Sheng-Po Huang

Wen-Jung Wu

Wei-Hsing Huang

Jing-Hong Wang

An input sequence re-ordering method with a multi input-precision reconfigurable scheme and a pipeline scheme for a computing-in-memory macro in a convolutional neural network application is configured to re-order a plurality of multi-bit input signals and includes performing a scanning step and a re-ordering step. The scanning step includes driving a scanner to scan one group of the multi-bit input signals to determine whether an initial value of a plurality of flag signals in one of a plurality of multi-bit section flags is changed to an inverted initial value according to a plurality of bit numbers of the one group of the multi-bit input signals. The re-ordering step includes driving a sorter to select a part of the one group of the multi-bit input signals corresponding to a plurality of the inverted initial values of the flag signals in the one of the multi-bit section flags.

Input sequence re-ordering method and input sequence re-ordering unit with multi input-precision reconfigurable scheme and pipeline scheme for computing-in-memory macro in convolutional neural network application

A package structure and the manufacturing method thereof are provided. The package structure includes a first package including at least one first semiconductor die encapsulated in an insulating encapsulation and through insulator vias electrically connected to the at least one first semiconductor die, a second package including at least one second semiconductor die and conductive pads electrically connected to the at least one second semiconductor die, and solder joints located between the first package and the second package. The through insulator vias are encapsulated in the insulating encapsulation. The first package and the second package are electrically connected through the solder joints. A maximum size of the solder joints is greater than a maximum size of the through insulator vias measuring along a horizontal direction, and is greater than or substantially equal to a maximum size of the conductive pads measuring along the horizontal direction.

Package structure and manufacturing method thereof

A dynamic gradient calibration method for a computing-in-memory neural network is performed to update a plurality of weights in a computing-in-memory circuit according to a plurality of inputs corresponding to a correct answer. A forward operating step includes performing a bit wise multiply-accumulate operation on a plurality of divided inputs and a plurality of divided weights to generate a plurality of multiply-accumulate values, and performing a clamping function on the multiply-accumulate values to generate a plurality of clamped multiply-accumulate values according to a predetermined upper bound value, and comparing the clamped multiply-accumulate values with the correct answer to generate a plurality of loss values. A backward operating step includes performing a partial differential operation on the loss values relative to the weights to generate a weight-based gradient. The weights are updated according to the weight-based gradient.

Dynamic gradient calibration method for computing-in-memory neural network and system thereof

An input-shaping method for a group-modulated input scheme in a plurality of computing-in-memory applications is configured to shape a plurality of multi-bit input signals. The input-shaping method for the group-modulated input scheme in the plurality of computing-in-memory applications includes performing an input splitting step, a threshold setting step and an input shaping step. The input splitting step includes splitting the multi-bit input signals into a plurality of input sub-groups via an input-shaping unit. The threshold setting step includes setting at least one shaping threshold via the input-shaping unit. The input shaping step includes shaping at least one of the input sub-groups according to the at least one shaping threshold via the input-shaping unit to form a plurality of shaped multi-bit input signals so as to increase a probability of a bit equal to 0 occurring in the at least one of the input sub-groups.

Input-shaping method and input-shaping unit for group-modulated input scheme in computing-in-memory applications

A quantization method for a plurality of partial sums of a convolution neural network based on a computing-in-memory hardware includes a probability-based quantizing step and a margin-based quantizing step. The probability-based quantizing step includes a network training step, a quantization-level generating step, a partial-sum quantizing step, a first network retraining step and a first accuracy generating step. The margin-based quantizing step includes a quantization edge changing step, a second network retraining step and a second accuracy generating step. The quantization edge changing step includes changing a quantization edge of at least one of a plurality of quantization levels. The probability-based quantizing step is performed to generate a first accuracy value, and the margin-based quantizing step is performed to generate a second accuracy value. The second accuracy value is greater than the first accuracy value.

Quantization method for partial sums of convolution neural network based on computing-in-memory hardware and system thereof

A method for controlling a touch system capable of reducing electromagnetic interference includes generating a first transmitting signal with a first frequency and a first alternating current shielding signal with the first frequency, transmitting the first transmitting signal to a first scan line of a plurality of scan lines of a touch panel, transmitting the first alternating current shielding signal to the plurality of scan lines exclusive of the first scan line, generating a second transmitting signal with a second frequency and a second alternating current shielding signal with the second frequency, transmitting the second transmitting signal to a second scan line of the plurality of scan lines, and transmitting the second alternating current shielding signal to the plurality of scan lines exclusive of the second scan line.

Touch system and method for controlling the touch system by varying frequencies of alternating current shielding signals and transmitting signals over time

A touch pad driving device and a touch driving method are provided. The touch pad driving device is configured to scan a plurality of sensing channels of a touch pad, cause the selected sensing channel in the plurality of sensing channels receive a touch driving signal, cause the sensing channels adjacent to or around the selected sensing channel receive a shielding signal, and cause the remaining sensing channels receive a reference signal, so as to reduce electromagnetic interference generated by the touch pad.

Touch pad driving device and touch driving method

The present invention concerns the uses of an azaphilone compound of formula (I): formula (I): or a pharmaceutically acceptable derivative thereof as described in the specification for modulation of the activity of a nuclear hormone receptor and for prevention and/or treatment of a disease or disorder related to nuclear hormone receptor activity.

Use of azaphilone compounds for the modulation of the activity of a nuclear hormone receptor

Growing community of inventors

Hsinchu, Taiwan

Ta-Wei Liu