Abstract: A reservoir device, comprises a first transistor and a second transistor. A gate of the first transistor is coupled to a write word line, a drain of the first transistor is coupled to a write bit line. A source of the second transistor is coupled to a read source line, a drain of the second transistor is coupled to a read bit line, and a gate of the second transistor is coupled to a source of the first transistor. A storage node is located on a coupling point between the gate of the second transistor and the source of the first transistor. The reservoir device selectively performs a write operation, a read operation or a refresh operation in response to an input voltage received by the write word line, the write bit line, the read source line and the read bit line respectively.

Abstract: A storage device for generating an identity code, includes a first storage circuit, a second storage circuit and a reading circuit. The first storage circuit stores several first data having several bits. The second storage circuit stores several second data having several bits. The reading circuit reads the second data from the second storage circuit to form a first sequence, and simultaneously reads the first data from the first storage circuit to form a second sequence. The reading circuit includes a processing circuit which simultaneously receives the first sequence and the second sequence, selects a first portion of the second sequence to form a target sequence according to the first sequence, and outputs the target sequence to serve as an identity code. Logical values of the bits of the first data and the second data are randomly distributed or pre-defined by a user.

Abstract: A hyperdimensional computing device includes a non-volatile memory cell array and a first operation circuit. The non-volatile memory cell array is coupled to a plurality of first word lines. The non-volatile memory cell array has a plurality of memory cell groups, a plurality of first memory cells of each of the memory cell groups are coupled to a same first word line of the first word lines, and the memory cell groups respectively store a plurality of data vectors. The first operation circuit receives at least one of the data vectors through bit lines and generates a bundled data vector according to the at least one of the data vectors.

Abstract: A memory device includes a first memory cell performing a logic operation. The first memory cell includes first and second switches. The first switch writes a first weight bit into a first storage node. The second switch generates a first current signal according to the first weight bit and a first input bit. The second switch receives a first bit line signal carrying the first input bit and a first word line signal. A control terminal of the second switch is coupled to the first storage node. When the first input bit has a first logic value, the first bit line signal and the first word line signal has a first voltage level. When the first input bit has a second logic value, the first bit line signal has a second voltage level smaller than the first voltage level.

Abstract: An artificial neural network operation circuit and an in-memory computation device of the artificial neural network operation circuit are proposed. The in-memory computation device includes a memory cell array, a compensation memory cell string, and an operator. The memory cell array has a plurality of memory cells to store a plurality of weight values. The memory cell array has a plurality of word lines and a plurality of bit lines. Each compensation memory cell of the compensation memory cell string stores a unit weight value. The operator multiplies a signal on a compensation bit line by peak weight information of the weight values to generate a first signal and adds the first signal to each signal on the bit lines to obtain a plurality of computation results, respectively.

Abstract: The present invention discloses a memory and a training method for neural network based on memory. The training method includes: obtaining one or more transfer functions of a memory corresponding to one or more influence factors; determining a training plan according to an ideal case and the one or more influence factors; training the neural network according to the training plan and the one or more transfer functions to obtain a plurality of weights of the trained neural network; and programming the memory according to the weights.

Abstract: A hybrid in-memory search (IMS) content addressable memory (CAM) cell includes: a first IMS CAM cell; and a second IMS CAM cell, coupled to the first IMS CAM cell. The first IMS CAM cell and the second IMS CAM cell are of different types. When the hybrid IMS CAM cell stores a storage data, the first IMS CAM cell stores a first part of the storage data and the second IMS CAM cell stores the storage data or a second part of the storage data.

Abstract: An in-dynamic memory search device and an operation method thereof are provided. The in-dynamic memory search device includes at least one word line, at least two bit lines, at least one match line, at least one unit cell, at least two search lines, at least one pre-charge unit and at least one sense unit. The unit cell includes two storage elements and two search transistors. Each of the storage elements includes a write transistor and a read transistor. The write transistor is connected to the word line and one of the bit lines. The read transistor is connected to the write transistor and the match line. The search transistors are respectively connected to the read transistors. The search lines are respectively connected to the search transistors. The pre-charge unit is connected to the match line. The sense unit is connected to the match line.

Abstract: The application provides a content addressable memory (CAM) device and a method for searching and comparing data thereof. The CAM device comprises: a plurality of memory strings; and a sensing amplifier circuit coupled to the memory strings; wherein in data searching, a search data is compared with a storage data stored in the memory strings, the memory strings generate a plurality of string currents, the sensing amplifier circuit senses the string currents to generate a plurality of sensing results; based on the sensing results, a match degree between the search data and the storage data is determined as one of the follows: all-matched, partially-matched and all-mismatched.

Abstract: The application discloses an integrated memory device, a manufacturing method and an operation method thereof. The integrated memory cell includes: a first memory cell; and an embedded second memory cell, serially coupled to the first memory cell, wherein the embedded second memory cell is formed on any one of a first side and a second side of the first memory cell.

Abstract: A memory device and associated operation method are provided. The operation method is applied to the memory device to determine whether a search input and in-memory data are matched. The memory device includes a memory array and a control circuit, and the memory array includes M*N memory cells. The operation method includes the following steps. A select voltage is applied to an n-th word line. A pass-through voltage is applied to (N?1) word lines. A first search voltage is applied to an m-th first bit-line, and a second search voltage is applied to an m-th second bit-line. An m-th first sensing current and an m-th second sensing current bit are selectively generated. Then, a sensing circuit in the control circuit generates a sensing circuit output. The sensing circuit output represents whether the m-th first sensing current and the m-th second sensing current are generated.

Abstract: A memory device and a method for manufacturing the memory device are provided. The memory device includes a stack and a plurality of memory strings. The stack is disposed on the substrate, and the stack includes a plurality of conductive layers and a plurality of insulating layers alternately stacked. The memory strings pass through the stack along a first direction, wherein a first memory string in the memory strings includes a first conductive pillar and a second conductive pillar, a channel layer, and a memory structure. The first conductive pillar and the second conductive pillar respectively extend along the first direction and are separated from each other. The channel layer is disposed between the first conductive pillar and the second conductive pillar. The memory structure surrounds the second conductive pillar, and the memory structure includes a resistive memory material.

Abstract: The application provides a content addressable memory (CAM) memory device, a CAM cell and a method for searching and comparing data thereof. The CAM device includes: a plurality of CAM cells; and an electrical characteristic detection circuit coupled to the CAM cells; wherein in data searching, a search data is compared with a storage data stored in the CAM cells, the CAM cells generate a plurality of memory cell currents, the electrical characteristic detection circuit detects the memory cell currents to generate a plurality of sensing results, or the electrical characteristic detection circuit detects a plurality of match line voltages on a plurality of match lines coupled to the CAM cells to generate the plurality of search results; and the storage data is a single-bit multi-level storage data and/or the search data is a single-bit multi-level search data.

Abstract: An artificial neural network operation circuit and an in-memory computation device of the artificial neural network operation circuit are proposed. The in-memory computation device includes a memory cell array, a compensation memory cell string, and an operator. The memory cell array has a plurality of memory cells to store a plurality of weight values. The memory cell array has a plurality of word lines and a plurality of bit lines. Each compensation memory cell of the compensation memory cell string stores a unit weight value. The operator multiplies a signal on a compensation bit line by peak weight information of the weight values to generate a first signal and adds the first signal to each signal on the bit lines to obtain a plurality of computation results, respectively.

Abstract: The application discloses an integrated memory device, a manufacturing method and an operation method thereof. The integrated memory cell includes: a first memory cell; and an embedded second memory cell, serially coupled to the first memory cell, wherein the embedded second memory cell is formed on any one of a first side and a second side of the first memory cell.

Abstract: A hyperdimensional computing device includes a non-volatile memory cell array and a first operation circuit. The non-volatile memory cell array is coupled to a plurality of first word lines. The non-volatile memory cell array has a plurality of memory cell groups, a plurality of first memory cells of each of the memory cell groups are coupled to a same first word line of the first word lines, and the memory cell groups respectively store a plurality of data vectors. The first operation circuit receives at least one of the data vectors through bit lines and generates a bundled data vector according to the at least one of the data vectors.

Abstract: An image capturing device includes a base, a light sensing element, an image capturing lens, and multiple light sources. The light sensing element is disposed on the base. The image capturing lens is disposed above the light sensing element. The light sources are disposed on the base and arranged beside the image capturing lens. Each of the light sources emits a light beam. A chief ray direction of the light beam has a horizontal component and a vertical component. The horizontal component and the vertical component are both greater than zero.

Abstract: An in-dynamic memory search device and an operation method thereof are provided. The in-dynamic memory search device includes at least one word line, at least two bit lines, at least one match line, at least one unit cell, at least two search lines, at least one pre-charge unit and at least one sense unit. The unit cell includes two storage elements and two search transistors. Each of the storage elements includes a write transistor and a read transistor. The write transistor is connected to the word line and one of the bit lines. The read transistor is connected to the write transistor and the match line. The search transistors are respectively connected to the read transistors. The search lines are respectively connected to the search transistors. The pre-charge unit is connected to the match line. The sense unit is connected to the match line.

Abstract: A hybrid in-memory search (IMS) content addressable memory (CAM) cell includes: a first IMS CAM cell; and a second IMS CAM cell, coupled to the first IMS CAM cell. The first IMS CAM cell and the second IMS CAM cell are of different types. When the hybrid IMS CAM cell stores a storage data, the first IMS CAM cell stores a first part of the storage data and the second IMS CAM cell stores the storage data or a second part of the storage data.

Abstract: A storage device for generating an identity code, includes a first storage circuit, a second storage circuit and a reading circuit. The first storage circuit stores several first data having several bits. The second storage circuit stores several second data having several bits. The reading circuit reads the second data from the second storage circuit to form a first sequence, and simultaneously reads the first data from the first storage circuit to form a second sequence. The reading circuit includes a processing circuit which simultaneously receives the first sequence and the second sequence, selects a first portion of the second sequence to form a target sequence according to the first sequence, and outputs the target sequence to serve as an identity code. Logical values of the bits of the first data and the second data are randomly distributed or pre-defined by a user.