A cognitive computer is a computer that hardwires

artificial intelligence Artificial intelligence (AI) is the capability of computer, computational systems to perform tasks typically associated with human intelligence, such as learning, reasoning, problem-solving, perception, and decision-making. It is a field of re ...

and

machine learning Machine learning (ML) is a field of study in artificial intelligence concerned with the development and study of Computational statistics, statistical algorithms that can learn from data and generalise to unseen data, and thus perform Task ( ...

algorithms into an

integrated circuit An integrated circuit (IC), also known as a microchip or simply chip, is a set of electronic circuits, consisting of various electronic components (such as transistors, resistors, and capacitors) and their interconnections. These components a ...

that closely reproduces the behavior of the human brain. It generally adopts a

neuromorphic engineering Neuromorphic computing is an approach to computing that is inspired by the structure and function of the human brain. A neuromorphic computer/chip is any device that uses physical artificial neurons to do computations. In recent times, the term ...

approach. Synonyms include neuromorphic chip and cognitive chip. In 2023, IBM's proof-of-concept NorthPole chip (optimized for 2-, 4- and 8-bit precision) achieved remarkable performance in

image recognition Computer vision tasks include methods for acquiring, processing, analyzing, and understanding digital images, and extraction of high-dimensional data from the real world in order to produce numerical or symbolic information, e.g. in the form o ...

. In 2013, IBM developed

Watson Scanning Habitable Environments with Raman and Luminescence for Organics and Chemicals (SHERLOC) is an ultraviolet Raman spectrometer that uses fine-scale imaging and an ultraviolet (UV) laser to determine fine-scale mineralogy, and detect orga ...

, a cognitive computer that uses

neural network A neural network is a group of interconnected units called neurons that send signals to one another. Neurons can be either biological cells or signal pathways. While individual neurons are simple, many of them together in a network can perfor ...

s and

deep learning Deep learning is a subset of machine learning that focuses on utilizing multilayered neural networks to perform tasks such as classification, regression, and representation learning. The field takes inspiration from biological neuroscience a ...

techniques. The following year, it developed the 2014

TrueNorth A cognitive computer is a computer that hardwires artificial intelligence and machine learning algorithms into an integrated circuit that closely reproduces the behavior of the human brain. It generally adopts a neuromorphic engineering approach. ...

microchip architecture which is designed to be closer in structure to the human brain than the

von Neumann architecture The von Neumann architecture—also known as the von Neumann model or Princeton architecture—is a computer architecture based on the '' First Draft of a Report on the EDVAC'', written by John von Neumann in 1945, describing designs discus ...

used in conventional computers. In 2017,

Intel Intel Corporation is an American multinational corporation and technology company headquartered in Santa Clara, California, and Delaware General Corporation Law, incorporated in Delaware. Intel designs, manufactures, and sells computer compo ...

also announced its version of a cognitive chip in "Loihi, which it intended to be available to university and research labs in 2018. Intel (most notably with its Pohoiki Beach and Springs systems),

Qualcomm Qualcomm Incorporated () is an American multinational corporation headquartered in San Diego, California, and Delaware General Corporation Law, incorporated in Delaware. It creates semiconductors, software and services related to wireless techn ...

, and others are improving neuromorphic processors steadily.

IBM TrueNorth chip

TrueNorth was a

neuromorphic Neuromorphic computing is an approach to computing that is inspired by the structure and function of the human brain. A neuromorphic computer/chip is any device that uses physical artificial neurons to do computations. In recent times, the term ...

CMOS Complementary metal–oxide–semiconductor (CMOS, pronounced "sea-moss ", , ) is a type of MOSFET, metal–oxide–semiconductor field-effect transistor (MOSFET) semiconductor device fabrication, fabrication process that uses complementary an ...

produced by

IBM International Business Machines Corporation (using the trademark IBM), nicknamed Big Blue, is an American Multinational corporation, multinational technology company headquartered in Armonk, New York, and present in over 175 countries. It is ...

in 2014. It is a

manycore processor Manycore processors are special kinds of multi-core processors designed for a high degree of parallel processing, containing numerous simpler, independent processor cores (from a few tens of cores to thousands or more). Manycore processors are use ...

network on a chip A network on a chip or network-on-chip (NoC or )This article uses the convention that "NoC" is pronounced . Therefore, it uses the convention "a" for the indefinite article corresponding to NoC ("a NoC"). Other sources may pronounce it as an ...

design, with 4096

core Core or cores may refer to: Science and technology * Core (anatomy), everything except the appendages * Core (laboratory), a highly specialized shared research resource * Core (manufacturing), used in casting and molding * Core (optical fiber ...

s, each one having 256 programmable simulated

neurons A neuron (American English), neurone (British English), or nerve cell, is an membrane potential#Cell excitability, excitable cell (biology), cell that fires electric signals called action potentials across a neural network (biology), neural net ...

for a total of just over a million neurons. In turn, each neuron has 256 programmable "

synapse In the nervous system, a synapse is a structure that allows a neuron (or nerve cell) to pass an electrical or chemical signal to another neuron or a target effector cell. Synapses can be classified as either chemical or electrical, depending o ...

s" that convey the signals between them. Hence, the total number of programmable synapses is just over 268 million (2²⁸). Its basic

transistor count The transistor count is the number of transistors in an electronic device (typically on a single substrate or silicon die). It is the most common measure of integrated circuit complexity (although the majority of transistors in modern microproc ...

is 5.4 billion.

Details

Memory, computation, and communication are handled in each of the 4096 neurosynaptic cores, TrueNorth circumvents the von Neumann-architecture bottleneck and is very energy-efficient, with IBM claiming a power consumption of 70

milliwatts The watt (symbol: W) is the unit of power or radiant flux in the International System of Units (SI), equal to 1 joule per second or 1 kg⋅m2⋅s−3. It is used to quantify the rate of energy transfer. The watt is named in honor of ...

and a power density that is 1/10,000th of conventional microprocessors. The

SyNAPSE In the nervous system, a synapse is a structure that allows a neuron (or nerve cell) to pass an electrical or chemical signal to another neuron or a target effector cell. Synapses can be classified as either chemical or electrical, depending o ...

chip operates at lower temperatures and power because it only draws power necessary for computation.

Skyrmion In particle theory, the skyrmion () is a topologically stable field configuration of a certain class of non-linear sigma models. It was originally proposed as a model of the nucleon by (and named after) Tony Skyrme in 1961. As a topological solito ...

s have been proposed as models of the synapse on a chip. The neurons are emulated using a Linear-Leak Integrate-and-Fire (LLIF) model, a simplification of the leaky

integrate-and-fire Biological neuron models, also known as spiking neuron models, are mathematical descriptions of the conduction of electrical signals in neurons. Neurons (or nerve cells) are electrically excitable cells within the nervous system, able to fire ...

model. According to IBM, it does not have a

clock A clock or chronometer is a device that measures and displays time. The clock is one of the oldest Invention, human inventions, meeting the need to measure intervals of time shorter than the natural units such as the day, the lunar month, a ...

, operates on unary numbers, and computes by counting to a maximum of 19 bits. The cores are event-driven by using both synchronous and asynchronous logic, and are interconnected through an asynchronous

packet-switched In telecommunications, packet switching is a method of grouping data into short messages in fixed format, i.e. '' packets,'' that are transmitted over a digital network. Packets consist of a header and a payload. Data in the header is used b ...

mesh network on chip (NOC). IBM developed a new network to program and use TrueNorth. It included a simulator, a new programming language, an integrated programming environment, and libraries. This lack of

backward compatibility In telecommunications and computing, backward compatibility (or backwards compatibility) is a property of an operating system, software, real-world product, or technology that allows for interoperability with an older legacy system, or with Input ...

with any previous technology (e.g., C++ compilers) poses serious

vendor lock-in In economics, vendor lock-in, also known as proprietary lock-in or customer lockin, makes a customer dependent on a vendor for products, unable to use another vendor without substantial switching costs. The use of open standards and alternati ...

risks and other adverse consequences that may prevent it from commercialization in the future.

Research

In 2018, a cluster of TrueNorth network-linked to a master computer was used in stereo vision research that attempted to extract the depth of rapidly moving objects in a scene.

IBM NorthPole chip

In 2023, IBM released its NorthPole chip, which is a

proof-of-concept A proof of concept (POC or PoC), also known as proof of principle, is an inchoate realization of a certain idea or method in order to demonstrate its feasibility or viability. A proof of concept is usually small and may or may not be complete ...

for dramatically improving performance by intertwining compute with memory on-chip, thus eliminating the

Von Neumann bottleneck The von Neumann architecture—also known as the von Neumann model or Princeton architecture—is a computer architecture based on the '' First Draft of a Report on the EDVAC'', written by John von Neumann in 1945, describing designs discus ...

. It blends approaches from IBM's 2014 TrueNorth system with modern hardware designs to achieve speeds about 4,000 times faster than TrueNorth. It can run ResNet-50 or Yolo-v4

tasks about 22 times faster, with 25 times less energy and 5 times less space, when compared to

GPUs A graphics processing unit (GPU) is a specialized electronic circuit designed for digital image processing and to accelerate computer graphics, being present either as a discrete video card or embedded on motherboards, mobile phones, personal ...

which use the same 12-nm node process that it was fabricated with. It includes 224 MB of

RAM Ram, ram, or RAM most commonly refers to: * A male sheep * Random-access memory, computer memory * Ram Trucks, US, since 2009 ** List of vehicles named Dodge Ram, trucks and vans ** Ram Pickup, produced by Ram Trucks Ram, ram, or RAM may also ref ...

and 256

processor core A central processing unit (CPU), also called a central processor, main processor, or just processor, is the primary Processor (computing), processor in a given computer. Its electronic circuitry executes Instruction (computing), instructions ...

s and can perform 2,048 operations per core per cycle at 8-bit precision, and 8,192 operations at 2-bit precision. It runs at between 25 and 425

MHz The hertz (symbol: Hz) is the unit of frequency in the International System of Units (SI), often described as being equivalent to one event (or cycle) per second. The hertz is an SI derived unit whose formal expression in terms of SI base u ...

. This is an inferencing chip, but it cannot yet handle GPT-4 because of memory and accuracy limitations

Intel Loihi chip

Pohoiki Springs

Pohoiki Springs is a system that incorporates Intel's self-learning neuromorphic chip, named Loihi, introduced in 2017, perhaps named after the Hawaiian

seamount A seamount is a large submarine landform that rises from the ocean floor without reaching the water surface (sea level), and thus is not an island, islet, or cliff-rock. Seamounts are typically formed from extinct volcanoes that rise abruptly a ...

Lōʻihi. Intel claims Loihi is about 1000 times more energy efficient than general-purpose computing systems used to train neural networks. In theory, Loihi supports both machine learning training and inference on the same silicon independently of a cloud connection, and more efficiently than

convolutional neural network A convolutional neural network (CNN) is a type of feedforward neural network that learns features via filter (or kernel) optimization. This type of deep learning network has been applied to process and make predictions from many different ty ...

s or

s. Intel points to a system for monitoring a person's heartbeat, taking readings after events such as exercise or eating, and using the chip to normalize the data and work out the ‘normal’ heartbeat. It can then spot abnormalities and deal with new events or conditions. The first iteration of the chip was made using Intel's 14 nm fabrication process and houses 128 clusters of 1,024

artificial neuron An artificial neuron is a mathematical function conceived as a model of a biological neuron in a neural network. The artificial neuron is the elementary unit of an ''artificial neural network''. The design of the artificial neuron was inspired ...

s each for a total of 131,072 simulated neurons. This offers around 130 million

s, far less than the human brain's 800

trillion ''Trillion'' is a number with two distinct definitions: *1,000,000,000,000, i.e. one million 1,000,000, million, or (ten to the twelfth Exponentiation, power), as defined on the long and short scales, short scale. This is now the meaning in bot ...

synapses, and behind IBM's

. Loihi is available for research purposes among more than 40 academic research groups as a

USB Universal Serial Bus (USB) is an industry standard, developed by USB Implementers Forum (USB-IF), for digital data transmission and power delivery between many types of electronics. It specifies the architecture, in particular the physical ...

form factor. In October 2019, researchers from

Rutgers University Rutgers University ( ), officially Rutgers, The State University of New Jersey, is a Public university, public land-grant research university consisting of three campuses in New Jersey. Chartered in 1766, Rutgers was originally called Queen's C ...

published a research paper to demonstrate the

energy efficiency Energy efficiency may refer to: * Energy efficiency (physics), the ratio between the useful output and input of an energy conversion process ** Electrical efficiency, useful power output per electrical power consumed ** Mechanical efficiency, a rat ...

of Intel's Loihi in solving

simultaneous localization and mapping Simultaneous localization and mapping (SLAM) is the computational problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an Intelligent agent, agent's location within it. While this initially ap ...

. In March 2020, Intel and

Cornell University Cornell University is a Private university, private Ivy League research university based in Ithaca, New York, United States. The university was co-founded by American philanthropist Ezra Cornell and historian and educator Andrew Dickson W ...

published a research paper to demonstrate the ability of Intel's Loihi to recognize different

hazardous materials Dangerous goods are substances that are a risk to health, safety, property or the Natural environment, environment during transport. Certain dangerous goods that pose risks even when not being transported are known as hazardous materials (syll ...

, which could eventually aid to "diagnose diseases, detect weapons and

explosive An explosive (or explosive material) is a reactive substance that contains a great amount of potential energy that can produce an explosion if released suddenly, usually accompanied by the production of light, heat, sound, and pressure. An ex ...

s, find

narcotic The term narcotic (, from ancient Greek ναρκῶ ''narkō'', "I make numb") originally referred medically to any psychoactive compound with numbing or paralyzing properties. In the United States, it has since become associated with opiates ...

s, and spot signs of smoke and

carbon monoxide Carbon monoxide (chemical formula CO) is a poisonous, flammable gas that is colorless, odorless, tasteless, and slightly less dense than air. Carbon monoxide consists of one carbon atom and one oxygen atom connected by a triple bond. It is the si ...

Pohoiki Beach

Intel's Loihi 2, named Pohoiki Beach, was released in September 2021 with 64 cores. It boasts faster speeds, higher-bandwidth inter-chip communications for enhanced scalability, increased capacity per chip, a more compact size due to process scaling, and improved programmability.

Hala Point

Hala Point packages 1,152 Loihi 2 processors produced on Intel 3 process node in a six-rack-unit chassis. The system supports up to 1.15 billion neurons and 128 billion synapses distributed over 140,544 neuromorphic processing cores, consuming 2,600 watts of power. It includes over 2,300 embedded x86 processors for ancillary computations. Intel claimed in 2024 that Hala Point was the world’s largest neuromorphic system. It uses Loihi 2 chips. It is claimed to offer 10x more neuron capacity and up to 12x higher performance. Hala Point provides up to 20 quadrillion operations per second, (20 petaops), with efficiency exceeding 15 trillion (8-bit) operations S⁻¹ W⁻¹ on conventional deep neural networks. Hala Point integrates processing, memory and communication channels in a massively parallelized fabric, providing 16 PB S⁻¹ of memory bandwidth, 3.5 PB S⁻¹ of inter-core communication bandwidth, and 5 TB S⁻¹ of inter-chip bandwidth. The system can process its 1.15 billion neurons 20 times faster than a human brain. Its neuron capacity is roughly equivalent to that of an

owl Owls are birds from the order Strigiformes (), which includes over 200 species of mostly solitary and nocturnal birds of prey typified by an upright stance, a large, broad head, binocular vision, binaural hearing, sharp talons, and feathers a ...

brain or the cortex of a

capuchin monkey The capuchin monkeys () are New World monkeys of the subfamily Cebinae. They are readily identified as the "Street organ, organ grinder" monkey, and have been used in many movies and television shows. The range of capuchin monkeys includes some t ...

. Loihi-based systems can perform inference and optimization using 100 times less energy at speeds as much as 50 times faster than CPU/GPU architectures. Intel claims that Hala Point can create LLMs but this has not been done. Much further research is needed

SpiNNaker

SpiNNaker A spinnaker is a sail designed specifically for sailing off the wind on courses between a Point of sail#Reaching, reach (wind at 90° to the course) to Point of sail#Running downwind, downwind (course in the same direction as the wind). Spinna ...

(Spiking Neural Network Architecture) is a

massively parallel Massively parallel is the term for using a large number of computer processors (or separate computers) to simultaneously perform a set of coordinated computations in parallel. GPUs are massively parallel architecture with tens of thousands of ...

manycore Manycore may refer to: * Manycore processor Manycore processors are special kinds of multi-core processors designed for a high degree of parallel processing, containing numerous simpler, independent processor cores (from a few tens of cores to th ...

supercomputer architecture Approaches to supercomputer architecture have taken dramatic turns since the earliest systems were introduced in the 1960s. Early supercomputer architectures pioneered by Seymour Cray relied on compact innovative designs and local parallelism to ...

Department of Computer Science, University of Manchester The Department of Computer Science at the University of Manchester is the longest established department of Computer Science in the United Kingdom and one of the largest. It is located in the Kilburn Building on the Wilmslow Road#Oxford Road, Oxfo ...

Criticism

Critics argue that a room-sized computer – as in the case of IBM's

– is not a viable alternative to a three-pound human brain. Some also cite the difficulty for a single system to bring so many elements together, such as the disparate sources of information as well as computing resources. In 2021, ''The New York Times'' released Steve Lohr's article "What Ever Happened to IBM’s Watson?". He wrote about some costly failures of IBM Watson. One of them, a cancer-related project called the Oncology Expert Advisor, was abandoned in 2016 as a costly failure. During the collaboration, Watson could not use patient data. Watson struggled to decipher doctors’ notes and patient histories.