Patent classifications
G11C11/005
Neuromorphic memory circuit and method of neurogenesis for an artificial neural network
A memory circuit configured to perform multiply-accumulate (MAC) operations for performance of an artificial neural network includes a series of synapse cells arranged in a cross-bar array. Each cell includes a memory transistor connected in series with a memristor. The memory circuit also includes input lines connected to the source terminal of the memory transistor in each cell, output lines connected to an output terminal of the memristor in each cell, and programming lines coupled to a gate terminal of the memory transistor in each cell. The memristor of each cell is configured to store a conductance value representative of a synaptic weight of a synapse connected to a neuron in the artificial neural network, and the memory transistor of each cell is configured to store a threshold voltage representative of a synaptic importance value of the synapse connected to the neuron in the artificial neural network.
Nonvolatile memory device and operation method thereof
A nonvolatile memory device includes a nonvolatile memory, a volatile memory being a cache memory of the nonvolatile memory, and a first controller configured to control the nonvolatile memory. The nonvolatile memory device further includes a second controller configured to receive a device write command and an address, and transmit, to the volatile memory through a first bus, a first read command and the address and a first write command and the address sequentially, and transmit a second write command and the address to the first controller through a second bus, in response to the reception of the device write command and the address.
Reconfigurable memory architectures
Techniques are described herein for a reconfigurable memory device that is configurable based on the type of interposer used to couple the memory device with a host device. The reconfigurable memory device may include a plurality components for a plurality of configurations. Various components of the reconfigurable memory die may be activated/deactivated based on what type of interposer is used in the memory device. For example, if a first type of interposer is used (e.g., a high-density interposer), the data channel may be eight data pins wide. In contrast, if second type of interposer is used (e.g., an organic-based interposer), the data channel may be four data pins wide. As such, a reconfigurable memory device may include data pins and related drivers that are inactive in some configurations.
Memory readout circuit and method
A circuit includes an array of OTP cells, an array of NVM cells, an amplifier coupled to each of the array of OTP cells and the array of NVM cells, and a control circuit configured to generate one or more control signals. Responsive to the one or more control signals, the amplifier is configured to generate an output voltage based on a current received from the array of OTP cells in a first configuration, and generate the output voltage based on a voltage received from the array of NVM cells in a second configuration.
Hardware accelerator with analog-content addressable memory (a-CAM) for decision tree computation
Examples described herein relate to a decision tree computation system in which a hardware accelerator for a decision tree is implemented in the form of an analog Content Addressable Memory (a-CAM) array. The hardware accelerator accesses a decision tree. The decision tree comprises of multiple paths and each path of the multiple paths includes a set of nodes. Each node of the decision tree is associated with a feature variable of multiple feature variables of the decision tree. The hardware accelerator combines multiple nodes among the set of nodes with a same feature variable into a combined single node. Wildcard values are replaced for feature variables not being evaluated in each path. Each combined single node associated with each feature variable is mapped to a corresponding column in the a-CAM array and the multiple paths of the decision tree to rows of the a-CAM array.
NON-VOLATILE MEMORY DEVICES AND SYSTEMS WITH VOLATILE MEMORY FEATURES AND METHODS FOR OPERATING THE SAME
Memory devices, systems including memory devices, and methods of operating memory devices and systems are provided, in which at least a subset of a non-volatile memory array is configured to behave as a volatile memory by erasing or degrading data in the event of a changed power condition such as a power-loss event, a power-off event, or a power-on event. In one embodiment of the present technology, a memory device is provided, comprising a non-volatile memory array, and circuitry configured to store one or more addresses of the non-volatile memory array, to detect a changed power condition of the memory device, and to erase or degrade data at the one or more addresses in response to detecting the changed power condition.
Neural network memory
An example apparatus can include a memory array and a memory controller. The memory array can include a first portion including a first plurality of memory cells. The memory array can further include a second portion including a second plurality of memory cells. The memory controller can be coupled to the first portion and the second portion. The memory controller can be configured to operate the first plurality of memory cells for short-term memory operations. The memory controller can be further configured to operate the second plurality of memory cells for long-term memory operations.
MEMORY SUB-SYSTEM REFRESH
A method includes determining a first memory access count threshold for a first word line of a block of memory cells and determining a second memory access count threshold for a second word line of the block of memory cells. The second memory access count threshold can be greater than the first memory access count threshold. The method can further include incrementing a memory block access count corresponding to the block of memory cells that includes the first word line and the second word line in response to receiving a memory access command and refreshing the first word line when the memory block access count corresponding to the block of memory cells is equal to the first memory access count threshold.
Ultra-low power neuromorphic artificial intelligence computing accelerator
A three-dimensional (3D) ultra-low power neuromorphic accelerator is described. The 3D ultra-low power neuromorphic accelerator includes a power manager as well as multiple tiers. The 3D ultra-low power neuromorphic accelerator also includes multiple cores defined on each tier and coupled to the power manager. Each core includes at least a processing element, a non-volatile memory, and a communications module.
Semiconductor assemblies including combination memory and methods of manufacturing the same
Semiconductor devices including vertically-stacked combination memory devices and associated systems and methods are disclosed herein. The vertically-stacked combination memory devices include at least one volatile memory die and at least one non-volatile memory die stacked on top of each other. The corresponding stack may be attached to a controller die that is configured to provide interface for the attached volatile and non-volatile memory dies.