10

H+ magazine wrote an estimate in 2009 that seems broadly comparable to other things I've seen; they think the human brain is approximately 37 petaflops. A supercomputer larger than that 37 petaflop estimate exists today. But emulation is hard. See this SO question about hardware emulation or this article on emulating the SNES, in which they require 140 ...


7

Quick Answer When Intel acquired Nirvana, they indicated their belief that analog VLSI has its place in the neuromorphic chips of the near future1, 2, 3. Whether it was because of the ability to more easily exploit the natural quantum noise in analog circuits is not yet public. It is more likely because of the number and complexity of parallel activation ...


6

In May 2016 Google announced a custom ASIC which was is specifically built for machine learningwiki and tailored for TensorFlow. It is using tensor processing unit (TPU) which is a programmable microprocessor designed to accelerate artificial neural networks. NeuroCores, 12x14 sq-mm chips which can be interconnected in a binary tree, see: Neurogrid, a ...


6

The human brain contains about 100 billion neurons ($10^{11}$) and about a hundred trillion synapses ($10^{14}$). Each neuron can fire about 100 times a second. If we model the brain as a simple neural network, then it would be equivalent to a machine that requires 1016 calculations per second and 1013 bits of memory. From Wikipedia Kurzweil introduces the ...


6

The approach you describe is called neuromorphic computing and it's quite a busy field. IBM's TrueNorth even has spiking neurons. The main problem with these projects is that nobody quite knows what to do with them yet. These projects don't try to create chips that are optimised to run a neural network. That would certainly be possible, but the ...


6

I think, there are various reasons. First of all: Flexibility. With modern CPUs and GPUs you can construct pretty much every AI model you want and in every size and complexity you want. How can you be sure that the model you are currently using is still suitable in a few years? Maybe there will be a major breakthrough in NNs in the next few years? Maybe some ...


5

Tensor operations The major work in most ML applications is simply a set of (very large) tensor operations e.g. matrix multiplication. You can do that easily in an ASIC, and all the other algorithms can just run on top of that.


4

If neurons and synapses can be implemented using transistors, I hope you are not talking about the neural networks which are currently winning all competitions in machine learning (MLPs, CNNs, RNNs, Deep Residual Networks, ...). Those were once used as a model for neurons, but they are only very loosely related to what happens in real brain cells. Spiking ...


3

Digital Instrumentation of the Analog Cells One of the key challenges in analog artificial networks is that network instrumentation would be most practical if digital. Any VLSI implementation of analog perceptrons, convolutions, or spiking networks will likely need to have digital components in a hybrid arrangement for several functions. Health indicators ...


3

Google has not released the manufacturing details for their TPUs. However, it's suspected that they're produced by either Taiwan Semiconductor Manufacturing or GlobalFoundries, as these are some of the largest companies in the industry. Source


3

I did some recent research on this topic. It all comes down to parallelization. Basically there are 2 ways to do it: model parallelization or batch parallelization. Model parallelization is when you split the model by layers among multiple GPUs. As per my best knowledge you can't split a layer between GPUs, so 8 GPUs would serve 8 layers that is very ...


3

Edge Computing is an approach for extended from cloud computing which leverages the same concept but has its advantage like mitigating the latency, resource usage, energy usage and so on. Federated learning is just an algorithm or a kind of approach which empower the edge computing by applying the technique of model iteration instead of fetching data from ...


3

Federated systems or Fog models, basically push computation from the active system side to server or network side processes. This is commonly used when computationally expensive services are required on limited systems (such as running some AI or augmentation occlusion processing from on phone), allowing for distributed data processing. Here is a good paper ...


3

No. YOLO and SSD are based on Nvidia's proprietary CUDA technology which is not available on Raspberry simply because of the GPU vendor is not Nvidia. Even more, there seems to be no implementation of even OpenCL for the Raspberry's GPU. What you can do is to try port YOLO's of SSD's CNN core from CUDA to Raspberry GPU's assembler, in the way described in, ...


3

I think the algorithm has changed minimally, but the necessary hardware has been trimmed to the bone. The number of gate transitions are reduced (perhaps float ops and precision too), as are the number of data move operations, thus saving both power and runtime. Google suggests their TPU achieves a 10X cost saving to get the same work done. https://...


3

AI or Artificial Intelligence is nothing but intelligence but in its artificial form. Intelligence comes from formation of rules and patterns in the data which is seen or on which it is trained. For us, we programme or formulate these rules in Mathematics on which an Intelligence is created. Ex. A neural network shows a lower level of Intelligence. It is ...


3

Neural networks are just software. Software is just one form of data in the Von Neumann Architecture. Most of the data is stored on the disk. There are two common types of disks: HDD: An actual disk which can be magnetified is spinning. SSD: Solid-state drive If the part where the data is stored is destroyed, the data is destroyed. Of course, the data can ...


3

First of all, I know GPU matters much more than CPU In general this is wrong or very naive. See also this answer (then consider SBCL) and that one (and also this and that). The OS I'm using is Windows 10, That is your biggest mistake, assuming you have access (and some understanding) to the source code of your application. See below why. Notice that the ...


3

XPU is a device abstraction for Intel heterogeneous computation architectures, which can be mapped to CPU, GPU, FPGA, and accelerator. In order to integrate a new accelerator you need 2 things: HW support: read about XPU in the official intel XPU webpage. SW support: see this open Feature Request to include this kind of architecture to the pytorch pool of ...


2

Specialized AI hardware takes advantage of highly parallelizable nature of many neural network designs. GPUs are designed for pixel crunching - coincidentally very paralelisable too. This is why they often offer orders of magnitude better performance (in NN tasks) than CPUs. That being said, GPUs have shortcomings to. First, many of GPU's built in features ...


2

Parallel processing is very important when computing ANNs. Though they are not designed for AIs, GPUs are widely used in machine learning because of their higher capability of paralel processing and it is due to their higher number of cores comparing CPUs. All in all, the dedicated hardware should have GPU-like architecture.


2

Theoretically, there shouldn't be a problem copying either of the artificial brains in any state. Difficulty in measuring a state doesn't seem to really be a problem until you get down to the quantum level, where the means of measurement affect the state. The configuration of the artificial brains, including pathway structures and states, should be ...


2

While a single transistor could approximate the basic function of a single neuron, I cannot agree that any electronic element could simulate the synapses/axons. Transistors are etched on a flat surface, and could be interconnected only to adjacent or close by transistors. Axons in the brain span huge distances (compared to the size of the neuron itself), and ...


2

Not just how much, but what kind of processing power : there're specially-crafted dedicated chips, and it has a practical applications, so it's not a lab-only project


2

The Mac Pro uses AMD GPUs. These don't support CUDA, but instead support the OpenCL framework. TensorFlow, the most popular deep learning library, uses CUDA to run on GPUs, although OpenCL support is in the works. That said, one of the main approaches to OpenCL support, SYCL, isn't planning to support OSX: We have no plans to support OSX in near future. ...


2

One can also approach the question from the information theory aspect: There are two trade/offs to choose from: Analog information that may represent information in a more precise/specific way, but limited in quantity. Digital information that doesn't fully represent the real world, but may contain unlimited amount of information within a few bits. A good ...


Only top voted, non community-wiki answers of a minimum length are eligible