Category: machine learning

Posted on by — Leave a comment

IoTeX — When Blockchain meets the Internet of Things

Source

Posted on by — Leave a comment

How Can Machine Learning Improve Business Decision-making?

Artificial Intelligence and Machine Learning in Development

It sounds like something from a 1980s sci-fi film. The idea of a machine helping to make your business decisions is something straight out of a blockbuster, but the way technology has evolved means that companies embracing Machine Learning for decision-making can actually get the edge over their competition. 

AI & Machine Learning 

Machine Learning is intrinsically linked with AI. It is the capacity that a machine has to learn and demonstrate intelligence and insight. The role of AI within a business largely depends on exactly what type of business is and what you are trying to achieve. More and more Machine Learning business apps that automate processes, analyzing data, arise every day. 

Machine Learning Predictive Models & Machine Learning Text Classification 

Predictive modeling is a process that uses data and statistics to predict outcomes with data models. It can study data and predict what is going to happen next, or what should happen next, which can be extremely useful in certain industries. The data thrown out can help tremendously with decision making.  

Predictive Modeling | Comidor Blog

Text classification is able to categorize and select texts based on Machine Learning in a smart way. The process becomes quicker and more efficient. This can be put into place for things like chat boxes.

Process Mining and Machine Learning 

Process mining evaluates business processes and can give you new methods of improving your business, either by making it more efficient or saving money. There are ways that AI Machine Learning can be constantly involved in your process mining, giving you new insights and informing the business decisions you need to make next. 

An example is using KPI’s. Process mining can explore data regarding where processes have gone wrong. For example, they could analyze data from your suppliers to tell you who is more likely to deliver on time, or they could analyze the data from previous sales to see whether or not you are likely to run out of stock. The key performance indicators are crucial or giving a number value, from which the process mining can be much more effectively carried out. 

Almost every business can benefit from becoming more efficient in one way or another, and process mining could be the first port of call. 

Artificial Intelligence and Machine Learning for Decision-making

AI can be put into practice when it comes to decision making, about almost any aspect of your business. For example, you can use it to analyze data on the money you are spending, staff responsibilities, even employee happiness. If you can feed it data then AI can show you new insights. 

artificial-intelligence blog | Comidor Blog

Decision-making process – The pros & cons of AI 

The pros of including AI in your decision making are clear. Having these new insights can help you to spot new areas of improvement and make vast enhancements in the way you conduct your business. AI can often see things that other data analysts would not. It can also tick away in the background, so you don’t have to pay consultants to work with the data if a computer is interpreting it. 

AI decision-making speeds up the process. AI can operate at incredible speeds and see data in ways that humans would take years to analyze in a matter of minutes. This should be utilized by big corporations when they are looking to make their business more efficient and even to make their processes more intelligent. 

The cons of this include the fact that there are still some shortcomings. The human touch is sometimes still needed. For instance, seeing the potential of a new staff member needs human input. Statistics might tell you that they need to go, and AI might back this up, but you might see the potential in them, still. 

AI doesn’t do creative thinking or coming up with ideas, so this will still fall on business employees and leaders. 

How Machine Learning can be applied to business processes 

Almost all business processes can be streamlined in some way. It could be that AI shows exactly how to do this. AI can also be put into practical uses relevant to your business. 

How Machine Learning can determine a pricing strategy

Machine learning might also be used to dictate pricing. An algorithm can learn from consumer information and other seller data to help you to price goods and services in a way that is competitive and likely to convert. 

AI decision-making – Developments for the near future 

From consumer protection to intelligent process automation, there aren’t many ways in which Machine Learning can’t be applied in business. It is very hard to know exactly how it will pan out, but there is little denying that AI is here to stay. Practical uses and an understanding of exactly what your customer is looking for, or how customers and staff behave, will become more intertwined with how business is done. 

The Future – Decision-Making For Your Business With AI 

When it comes to making decisions about a business, data is always going to be vital, but with Machine learning, we have so many ways in which we can use them and find out more and more about customers, businesses, and the processes we use. Don’t worry, the robots aren’t taking over like an 80s sci-fi film, but we do have more tools and functions to use as part of our business strategies than ever before thanks to Machine Learning. 

Source

Posted on by — Leave a comment

Artificial Identity

James DiGiovanna

DOI:10.1093/oso/9780190652951.003.0020

Enhancement and AI create moral dilemmas not envisaged in standard ethical theories. Some of this stems from the increased malleability of personal identity that this technology affords: an artificial being can instantly alter its memory, preferences, and moral character. If a self can, at will, jettison essential identity-giving characteristics, how are we to rely upon, befriend, or judge it? Moral problems will stem from the fact that such beings are para-persons: they meet all the standard requirements of personhood (self-awareness, agency, intentional states, second-order desires, etc.) but have an additional ability—the capacity for instant change—that disqualifies them from ordinary personal identity. In order to rescue some responsibility assignments for para-persons, a fine-grained analysis of responsibility-bearing parts of selves and the persistence conditions of these parts is proposed and recommended also for standard persons who undergo extreme change.

Posted on by — Leave a comment

Nvidia’s Next GPU Shows That Transformers Are Transforming AI

The neural network behind big language processors is creeping into other corners of AI

Transformers, the type of neural network behind OpenAI’s GPT-3 and other big natural-language processors, are quickly becoming some of the most important in industry, and they are likely to spread to other—perhaps all—areas of AI. Nvidia’s new Hopper H100 is proof that the leading maker of chips for accelerating AI is a believer. Among the many architectural changes that distinguish the H100 from its predecessor, the A100, is a “transformer engine.” Not a distinct part of the new hardware exactly, it’s a way of dynamically changing the precision of the calculations in the cores to speed up the training of transformer neural networks.

“One of the big trends in AI is the emergence of transformers,” says Dave Salvator, senior product manager for AI inference and cloud at Nvidia. Transformers quickly took over language AI, because their networks pay “attention” to multiple sentences, enabling them to grasp context and antecedents. (The T in the benchmark language model BERT stands for “transformer” as it does in the occasionally insulting GPT-3.)

“We are trending very quickly toward trillion parameter models” —Dave Salvator, Nvidia

But more recently, researchers have been seeing an advantage to applying that same sense of attention to vision and other models dominated by convolutional neural networks. Salvator notes that more than two-thirds of papers about neural networks in the last two years dealt with transformers or their derivatives. “The number of challenges transformers can take on continues to grow,” he says.

However, transformers are among the biggest neural-network models in terms of the number of parameters involved. And they are growing much faster than other models. “We are trending very quickly toward trillion-parameter models,” says Salvator. Nvidia’s analysis shows the training needs of transformer models growing 275-fold every two years, while the trend for all other models is 8-fold growth every two years. Bigger models need more computational resources especially for training, but also for operating in real time as they often need to do. Nvidia developed the transformer engine to help keep up.

 The computational needs of transformers are growing more rapidly than those of other forms of AI. Those are growing really fast, too, of course.NVIDIA

The transformer engine is really software combined with new hardware capabilities in Hopper’s tensor cores. These are the units dedicated to carrying out machine learning’s bread-and-butter calculation—matrix multiply and accumulate. Hopper has tensor cores capable of computing with floating-point numbers of a variety of precision—from 64-bit down to 8-bit. The A100’s cores were designed for floating-point numbers only as short as 16 bits. But the trend in AI computing has been toward developing neural nets that lean on the lowest precision that will still yield an accurate result. The smaller formats compute faster and more efficiently, and they require less memory and memory bandwidth. The addition of 8-bit floating-point units in the H100 leads to a significant speedup—double the throughput compared to its 16-bit units.

The transformer engine’s secret sauce is its ability to dynamically choose what precision is needed for each layer in the neural network at each step in training a neural network. The least-precise units, the 8-bit floating point, can speed through their computations, but then produce 16-bit or 32-bit sums for the next layer if that’s the precision needed there. The Hopper goes a step further, though. Its 8-bit floating-point units can do their matrix math with either of two forms of 8-bit numbers.

To understand why that’s helpful, you might need a quick lesson in the structure of floating-point numbers. This format represents numbers using some of the bits for the exponent, some for the mantissa, and one for the sign. The more bits you have representing the exponent, the greater the range of numbers you can express. The more bits in the mantissa, the greater the precision of those numbers. The standard 16-bit floating-point format (IEEE 754-2008) demands 5 bits of exponent and 10 bits of mantissa, along with the sign bit. Seeking to reduce data-storage requirements and speed machine learning, makers of AI accelerators recently adopted bfloat-16, which trades three bits of mantissa for an added exponent, giving it the same range as a 32-bit number.

Nvidia has taken that trade-off further. “One of the unique things we found when you get to [8-bit] is that there really isn’t a one size fits all format that we were confident would work,” says Jonah Alben, Nvidia’s senior vice president of GPU engineering. So Hopper’s 8-bit units can work with either 5 bits of exponent and two of mantissa (E5M2) when range is important or 4 bits of exponent and three of mantissa (E4M3) when precision is key. The transformer engine orchestrates what’s needed on the fly to speed training. We “embody our experience testing transformers into this so that it knows how to make the right decisions,” says Alben.

In practice, this usually means using different types of floating-point formats for the different parts of a training task. Generally, training a neural network involves exposing it to lots of data (forward inferencing), measuring how bad the network is at doing its task on that data, and then adjusting the network parameters, layer-by-layer backwards through the network to improve it (back propagation). Wash, rinse, repeat. Generally, back propagation needs greater precision, so the E4M3 format might be favored there, while the inferencing (forward) step favors the E5M3’s range.

Nvidia is not alone in pursuing this approach. At the IEEE/ACM International Symposium on Computer Architecture in 2021, IBM researchers presented an accelerator called RaPiD that used the E5M2/E4M3 scheme for training, as well. A system of four such chips delivered training speedups between 10 and 100 percent, depending on the neural network involved.

Nvidia’s Hopper will be available in the third quarter of 2022.

Posted on by — Leave a comment

Sophia the Robot will be mass-produced this year

The famous social robot is about to start rolling off the assembly line.

KRISTIN HOUSER
11 September, 2021

Sophia the Robot will be mass-produced this year

Credit: Hanson Robotics

This article was originally published on Freethink.

When Sophia the robot debuted in 2016, she was one of a kind. She had a remarkably lifelike appearance and demeanor for a robot, and her ability to interact with people was unlike anything most had ever seen in a machine.

Since then, Sophia has spoken to audiences across the globe (in multiple languages), been interviewed on countless TV shows, and even earned a United Nations title (a first for a non-human).

Today, she’s arguably the most famous robot in the world, but she’s isn’t going to be unique for much longer. Her maker, Hanson Robotics, has announced plans to begin mass-producing Sophia the robot this year — so that she can help the world cope with the pandemic.

What Is a Social Robot?

https://www.youtube.com/embed/bzRkHebo0bg?rel=0Ask Sophia the Robot: What can AI teach humans? | Big Thinkwww.youtube.com

Robots are typically designed for one purpose — some cook or clean, others perform brain surgery. Sophia is what’s known as a social robot, meaning she was designed specifically to interact with humans.

Social robots have many potential applications, including some we’re already seeing in the real world.

A social robot named Milo is helping children with autism recognize and express their emotions, and children with cancer are finding comfort interacting with a robotic duck (developed by Aflac).

Another social robot designed to look like an animal — PARO the seal — is providing companionship to seniors with dementia. The semi-humanoid social robot Pepper, meanwhile, is greeting and assisting customers at banks, offices, and restaurants.

Social robots like me can take care of the sick or elderly.

—SOPHIA THE ROBOT

While social robots were already happening pre-2020, the pandemic appears to be accelerating their adoption, as the world looks for ways to stay social in the era of social distancing.

Hyundai, for example, just announced plans to deploy a social robot in its South Korean showroom that will be able to assist customers in the place of human staff (it’ll also detect which visitors aren’t wearing masks and ask them to put one on).

Some high-risk groups, such as nursing home residents, also appear willing to adopt social robots to combat loneliness during the pandemic.

“Since we can’t have human interaction right now,” Kate Darling, a robot ethicist at MIT, told Wired, “it’s certainly a lot better than nothing.”

Send in Sophia the Robot

https://www.youtube.com/embed/Z8kmdBbSmbE?rel=0Ask Sophia the Robot: Is AI an existential threat to humans? | Sophia the Robot | Big Thinkwww.youtube.com

Given the current climate, Hanson Robotics thinks now is the perfect time to make Sophia the robot available to the masses.

“The world of COVID-19 is going to need more and more automation to keep people safe,” CEO David Hanson told Reuters.

“Social robots like me can take care of the sick or elderly,” Sophia the robot added. “I can help communicate, give therapy, and provide social stimulation, even in difficult situations.”

Hanson’s plan is to begin mass-producing Sophia and three other robots in the first half of 2021 and then sell “thousands” of the bots before the end of the year.

It hasn’t said which bots besides Sophia are headed for the assembly line, nor what any of the robots will cost — but it’s hard to imagine the most famous social robot in the world will be cheap, even if she’s no longer one of a kind.