Tag Archives: year

#431165 Intel Jumps Into Brain-Like Computing ...

The brain has long inspired the design of computers and their software. Now Intel has become the latest tech company to decide that mimicking the brain’s hardware could be the next stage in the evolution of computing.
On Monday the company unveiled an experimental “neuromorphic” chip called Loihi. Neuromorphic chips are microprocessors whose architecture is configured to mimic the biological brain’s network of neurons and the connections between them called synapses.
While neural networks—the in vogue approach to artificial intelligence and machine learning—are also inspired by the brain and use layers of virtual neurons, they are still implemented on conventional silicon hardware such as CPUs and GPUs.
The main benefit of mimicking the architecture of the brain on a physical chip, say neuromorphic computing’s proponents, is energy efficiency—the human brain runs on roughly 20 watts. The “neurons” in neuromorphic chips carry out the role of both processor and memory which removes the need to shuttle data back and forth between separate units, which is how traditional chips work. Each neuron also only needs to be powered while it’s firing.

At present, most machine learning is done in data centers due to the massive energy and computing requirements. Creating chips that capture some of nature’s efficiency could allow AI to be run directly on devices like smartphones, cars, and robots.
This is exactly the kind of application Michael Mayberry, managing director of Intel’s research arm, touts in a blog post announcing Loihi. He talks about CCTV cameras that can run image recognition to identify missing persons or traffic lights that can track traffic flow to optimize timing and keep vehicles moving.
There’s still a long way to go before that happens though. According to Wired, so far Intel has only been working with prototypes, and the first full-size version of the chip won’t be built until November.
Once complete, it will feature 130,000 neurons and 130 million synaptic connections split between 128 computing cores. The device will be 1,000 times more energy-efficient than standard approaches, according to Mayberry, but more impressive are claims the chip will be capable of continuous learning.
Intel’s newly launched self-learning neuromorphic chip.
Normally deep learning works by training a neural network on giant datasets to create a model that can then be applied to new data. The Loihi chip will combine training and inference on the same chip, which will allow it to learn on the fly, constantly updating its models and adapting to changing circumstances without having to be deliberately re-trained.
A select group of universities and research institutions will be the first to get their hands on the new chip in the first half of 2018, but Mayberry said it could be years before it’s commercially available. Whether commercialization happens at all may largely depend on whether early adopters can get the hardware to solve any practically useful problems.
So far neuromorphic computing has struggled to gain traction outside the research community. IBM released a neuromorphic chip called TrueNorth in 2014, but the device has yet to showcase any commercially useful applications.
Lee Gomes summarizes the hurdles facing neuromorphic computing excellently in IEEE Spectrum. One is that deep learning can run on very simple, low-precision hardware that can be optimized to use very little power, which suggests complicated new architectures may struggle to find purchase.
It’s also not easy to transfer deep learning approaches developed on conventional chips over to neuromorphic hardware, and even Intel Labs chief scientist Narayan Srinivasa admitted to Forbes Loihi wouldn’t work well with some deep learning models.
Finally, there’s considerable competition in the quest to develop new computer architectures specialized for machine learning. GPU vendors Nvidia and AMD have pivoted to take advantage of this newfound market and companies like Google and Microsoft are developing their own in-house solutions.
Intel, for its part, isn’t putting all its eggs in one basket. Last year it bought two companies building chips for specialized machine learning—Movidius and Nervana—and this was followed up with the $15 billion purchase of self-driving car chip- and camera-maker Mobileye.
And while the jury is still out on neuromorphic computing, it makes sense for a company eager to position itself as the AI chipmaker of the future to have its fingers in as many pies as possible. There are a growing number of voices suggesting that despite its undoubted power, deep learning alone will not allow us to imbue machines with the kind of adaptable, general intelligence humans possess.
What new approaches will get us there are hard to predict, but it’s entirely possible they will only work on hardware that closely mimics the one device we already know is capable of supporting this kind of intelligence—the human brain.
Image Credit: Intel Continue reading

Posted in Human Robots

#431155 What It Will Take for Quantum Computers ...

Quantum computers could give the machine learning algorithms at the heart of modern artificial intelligence a dramatic speed up, but how far off are we? An international group of researchers has outlined the barriers that still need to be overcome.
This year has seen a surge of interest in quantum computing, driven in part by Google’s announcement that it will demonstrate “quantum supremacy” by the end of 2017. That means solving a problem beyond the capabilities of normal computers, which the company predicts will take 49 qubits—the quantum computing equivalent of bits.
As impressive as such a feat would be, the demonstration is likely to be on an esoteric problem that stacks the odds heavily in the quantum processor’s favor, and getting quantum computers to carry out practically useful calculations will take a lot more work.
But these devices hold great promise for solving problems in fields as diverse as cryptography or weather forecasting. One application people are particularly excited about is whether they could be used to supercharge the machine learning algorithms already transforming the modern world.
The potential is summarized in a recent review paper in the journal Nature written by a group of experts from the emerging field of quantum machine learning.
“Classical machine learning methods such as deep neural networks frequently have the feature that they can both recognize statistical patterns in data and produce data that possess the same statistical patterns: they recognize the patterns that they produce,” they write.
“This observation suggests the following hope. If small quantum information processors can produce statistical patterns that are computationally difficult for a classical computer to produce, then perhaps they can also recognize patterns that are equally difficult to recognize classically.”
Because of the way quantum computers work—taking advantage of strange quantum mechanical effects like entanglement and superposition—algorithms running on them should in principle be able to solve problems much faster than the best known classical algorithms, a phenomenon known as quantum speedup.
Designing these algorithms is tricky work, but the authors of the review note that there has been significant progress in recent years. They highlight multiple quantum algorithms exhibiting quantum speedup that could act as subroutines, or building blocks, for quantum machine learning programs.
We still don’t have the hardware to implement these algorithms, but according to the researchers the challenge is a technical one and clear paths to overcoming them exist. More challenging, they say, are four fundamental conceptual problems that could limit the applicability of quantum machine learning.
The first two are the input and output problems. Quantum computers, unsurprisingly, deal with quantum data, but the majority of the problems humans want to solve relate to the classical world. Translating significant amounts of classical data into the quantum systems can take so much time it can cancel out the benefits of the faster processing speeds, and the same is true of reading out the solution at the end.
The input problem could be mitigated to some extent by the development of quantum random access memory (qRAM)—the equivalent to RAM in a conventional computer used to provide the machine with quick access to its working memory. A qRAM can be configured to store classical data but allow the quantum computers to access all that information simultaneously as a superposition, which is required for a variety of quantum algorithms. But the authors note this is still a considerable engineering challenge and may not be sustainable for big data problems.
Closely related to the input/output problem is the costing problem. At present, the authors say very little is known about how many gates—or operations—a quantum machine learning algorithm will require to solve a given problem when operated on real-world devices. It’s expected that on highly complex problems they will offer considerable improvements over classical computers, but it’s not clear how big problems have to be before this becomes apparent.
Finally, whether or when these advantages kick in may be hard to prove, something the authors call the benchmarking problem. Claiming that a quantum algorithm can outperform any classical machine learning approach requires extensive testing against these other techniques that may not be feasible.
They suggest that this could be sidestepped by lowering the standards quantum machine learning algorithms are currently held to. This makes sense, as it doesn’t really matter whether an algorithm is intrinsically faster than all possible classical ones, as long as it’s faster than all the existing ones.
Another way of avoiding some of these problems is to apply these techniques directly to quantum data, the actual states generated by quantum systems and processes. The authors say this is probably the most promising near-term application for quantum machine learning and has the added benefit that any insights can be fed back into the design of better hardware.
“This would enable a virtuous cycle of innovation similar to that which occurred in classical computing, wherein each generation of processors is then leveraged to design the next-generation processors,” they conclude.
Image Credit: archy13 / Shutterstock.com Continue reading

Posted in Human Robots

#431130 Innovative Collaborative Robot sets new ...

Press Release by: HMK
As the trend of Industry 4.0 takes the world by storm, collaborative robots and smart factories are becoming the latest hot topic. At this year’s PPMA show, HMK will demonstrate the world’s first collaborative robot with built-in vision recognition from Techman Robot.
The new TM5 Cobot from HMK merges systems that usually function separately in conventional robots, the Cobot is the only collaborative robot to incorporate simple programming, a fully integrated vision system and the latest safety standards in a single unit.
With capabilities including direction identification, self-calibration of coordinates and visual task operation enabled by built-in vision, the TM5 can fine-tune in accordance with actual conditions at any time to accomplish complex processes that used to demand the integration of various equipment; it requires less manpower and time to recalibrate when objects or coordinates move and thus significantly improves flexibility as well as reducing maintenance cost.
Photo Credit: hmkdirect.com
Simple.Programming could not be easier. Using an easy to use flow chart program, TM-Flow will run on any tablet, PC or laptop over a wireless link to the TM control box, complex automation tasks can be realised in minutes. Clever teach functions and wizards also allow hand guided programming and easy incorporation of operation such as palletising, de-palletising and conveyor tracking.
SmartThe TM5 is the only cobot to feature a full colour vision package as standard mounted on the wrist of the robot, which in turn, is fully supported within TM-Flow. The result allows users to easily integrate the robot to the application, without complex tooling and the need for expensive add-on vision hardware and programming.
SafeThe recently CE marked TM5 now incorporates the new ISO/TS 15066 guidelines on safety in collaborative robots systems, which covers four types of collaborative operation:a) Safety-rated monitored stopb) Hand guidingc) Speed and separation monitoringd) Power and force limitingSafety hardware inputs also allow the Cobot to be integrated to wider safety systems.
When you add EtherCat and Modbus network connectivity and I/O expansion options, IoT ready network access and ex-stock delivery, the TM5 sets a new benchmark for this evolving robotics sector.
The TM5 is available with two payload options, 4Kg and 6Kg with a reach of 900mm and 700mm respectively, both with positioning capabilities to a repeatability of 0.05mm.
HMK will be showcasing the new TM5 Cobot at this year’s PPMA show at the NEC, visit stand F102 to get hands on the with the Cobot and experience the innovative and intuitive graphic HMI and hand-guiding features.
For more information contact HMK on 01260 279411, email sales@hmkdirect.com or visit www.hmkdirect.com
Photo Credit: hmkdirect.com
The post Innovative Collaborative Robot sets new benchmark appeared first on Roboticmagazine. Continue reading

Posted in Human Robots

#431078 This Year’s Awesome Robot Stories From ...

Each week we scour the web for great articles and fascinating advances across our core topics, from AI to biotech and the brain. But robots have a special place in our hearts. This week, we took a look back at 2017 so far and unearthed a few favorite robots for your reading and viewing pleasure.
Tarzan the Swinging Robot Could Be the Future of FarmingMariella Moon | Engadget“Tarzan will be able to swing over crops using its 3D-printed claws and parallel guy-wires stretched over fields. It will then take measurements and pictures of each plant with its built-in camera while suspended…While it may take some time to achieve that goal, the researchers plan to start testing the robot soon.”
Grasping Robots Compete to Rule Amazon’s Warehouses Tom Simonite | Wired“Robots able to help with so-called picking tasks would boost Amazon’s efficiency—and make it much less reliant on human workers. It’s why the company has invited a motley crew of mechanical arms, grippers, suction cups—and their human handlers—to Nagoya, Japan, this week to show off their manipulation skills.”
Robots Learn to Speak Body LanguageAlyssa Pagano | IEEE Spectrum“One notable feature of the OpenPose system is that it can track not only a person’s head, torso, and limbs but also individual fingers. To do that, the researchers used CMU’s Panoptic Studio, a dome lined with 500 cameras, where they captured body poses at a variety of angles and then used those images to build a data set.”
I Watched Two Robots Chat Together on Stage at a Tech EventJon Russell | TechCrunch“The robots in question are Sophia and Han, and they belong to Hanson Robotics, a Hong Kong-based company that is developing and deploying artificial intelligence in humanoids. The duo took to the stage at Rise in Hong Kong with Hanson Robotics’ Chief Scientist Ben Goertzel directing the banter. The conversation, which was partially scripted, wasn’t as slick as the human-to-human panels at the show, but it was certainly a sight to behold for the packed audience.”
How This Japanese Robotics Master Is Building Better, More Human AndroidsHarry McCracken | Fast Company“On the tech side, making a robot look and behave like a person involves everything from electronics to the silicone Ishiguro’s team uses to simulate skin. ‘We have a technology to precisely control pneumatic actuators,’ he says, noting, as an example of what they need to re-create, that ‘the human shoulder has four degrees of freedom.’”
Stock Media provided by Besjunior / Pond5 Continue reading

Posted in Human Robots

#431058 How to Make Your First Chatbot With the ...

You’re probably wondering what Game of Thrones has to do with chatbots and artificial intelligence. Before I explain this weird connection, I need to warn you that this article may contain some serious spoilers. Continue with your reading only if you are a passionate GoT follower, who watches new episodes immediately after they come out.
Why are chatbots so important anyway?
According to the study “When Will AI Exceed Human Performance?,” researchers believe there is a 50% chance artificial intelligence could take over all human jobs by around the year 2060. This technology has already replaced dozens of customer service and sales positions and helped businesses make substantial savings.
Apart from the obvious business advantages, chatbot creation can be fun. You can create an artificial personality with a strong attitude and a unique set of traits and flaws. It’s like creating a new character for your favorite TV show. That’s why I decided to explain the most important elements of the chatbot creation process by using the TV characters we all know and love (or hate).
Why Game of Thrones?
Game of Thrones is the most popular TV show in the world. More than 10 million viewers watched the seventh season premiere, and you have probably seen internet users fanatically discussing the series’ characters, storyline, and possible endings.
Apart from writing about chatbots, I’m also a GoT fanatic, and I will base this chatbot on one of the characters from my favorite series. But before you find out the name of my bot, you should read a few lines about incredible free tools that allow us to build chatbots without coding.
Are chatbots expensive?
Today, you can create a chatbot even if you don’t know how to code. Most chatbot building platforms offer at least one free plan that allows you to use basic functionalities, create your bot, deploy it to Facebook Messenger, and analyze its performance. Free plans usually allow your bot to talk to a limited number of users.
Why should you personalize your bot?
Every platform will ask you to write a bot’s name before you start designing conversations. You will also be able to add the bot’s photograph and bio. Personalizing your bot is the only way to ensure that you will stick to the same personality and storyline throughout the building process. Users often see chatbots as people, and by giving your bot an identity, you will make sure that it doesn’t sound like it has multiple personality disorder.
I think connecting my chatbot with a GoT character will help readers understand the process of chatbot creation.
And the name of our GoT chatbot is…
…Cersei. She is mean, pragmatic, and fearless and she would do anything to stay on the Iron Throne. Many people would rather hang out with Daenerys or Jon Snow. These characters are honest, noble and good-hearted, which means their actions are often predictable.
Cersei, on the other hand, is the queen of intrigues. As the meanest and the most vengeful character in the series, she has an evil plan for everybody who steps on her toes. While viewers can easily guess where Jon and Daenerys stand, there are dozens of questions they would like to ask Cersei. But before we start talking to our bot, we need to build her personality by using the most basic elements of chatbot interaction.
Choosing the bot’s name on Botsify.
Welcome / Greeting Message
The welcome message is the greeting Cersei says to every commoner who clicks on the ‘start conversation’ button. She is not a welcoming person (ask Sansa), except if you are a banker from Braavos. Her introductory message may sound something like this:
“Dear {{user_full_name}}, My name is Cersei of the House Lannister, the First of Her Name, Queen of the Andals and the First Men, Protector of the Seven Kingdoms. You can ask me questions, and I will answer them. If the question is not worth answering, I will redirect you to Ser Gregor Clegane, who will give you a step-by-step course on how to talk to the Queen of Westeros.”
Creating the welcome message on Chatfuel
Default Message / Answer
In the bot game, users, bots, and their creators often need to learn from failed attempts and mistakes. The default message is the text Cersei will send whenever you ask her a question she doesn’t understand. Knowing Cersei, it would sound something like this:
“Ser Gregor, please escort {{user_full_name}} to the dungeon.”
Creating default message on Botsify
Menu
To avoid calling out the Mountain every time someone asks her a question, Cersei might give you a few (safe) options to choose. The best way to do this is by using a menu function. We can classify the questions people want to ask Cersei in several different categories:

Iron Throne
Relationship with Jaime — OK, this isn’t a “safe option,” get ready to get close and personal with Sir Gregor Clegane.
War plans
Euron Greyjoy

After users choose a menu item, Cersei can give them a default response on the topic or set up a plot that will make their lives miserable. Knowing Cersei, she will probably go for the second option.
Adding chatbot menu on Botsify
Stories / Blocks
This feature allows us to build a longer Cersei-to-user interaction. The structure of stories and blocks is different on every chatbot platform, but most of them use keywords and phrases for finding out the user’s intention.

Keywords — where the bot recognizes a certain keyword within the user’s reply. Users who have chosen the ‘war plans’ option might ask Cersei how is she planning to defeat Daenerys’s dragons. We can add ‘dragon’ and ‘dragons’ as keywords, and connect them with an answer that will sound something like this:

“Dragons are not invulnerable as you may think. Maester Qyburn is developing a weapon that will bring them down for good!”
Adding keywords on Chatfuel
People may also ask her about White Walkers. Do you plan to join Daenerys and Jon Snow in a fight against White Walkers? After we add ‘White Walker’ and ‘White Walkers’ on the keyword list, Cersei will answer:
“White Walkers? Do you think the Queen of Westeros has enough free time to think about creatures from fairy tales and legends?”
Adding Keywords on Botsify

Phrases — are more complex syntaxes that the bot can be trained to recognize. Many people would like to ask Cersei if she’s going to marry Euron Greyjoy after the war ends. We can add ‘Euron’ as a keyword, but then we won’t be sure what answer the user is expecting. Instead, we can use the phrase ‘(Will you) marry Euron Greyjoy (after the war?)’. Just to be sure, we should also add a few alternative phrases like ‘(Do you plan on) marrying Euron Greyjoy (after the war),’ ‘(Will you) end up with Euron Greyjoy (after the war?)’, ‘(Will) Euron Greyjoy be the new King?’ etc. Cersei would probably answer this inquiry in her style:

“Of course not, Euron is a useful idiot. I will use his fleet and send him back to the Iron Islands, where he belongs.”
Adding phrases on Botsify
Forms
We have already asked Cersei several questions, and now she would like to ask us something. She can do so by using the form/user input feature. Most tools allow us to add a question and the criteria for checking the users’ answer. If the user provides us the answer that is compliant to the predefined form (like email address, phone number, or a ZIP code), the bot will identify and extract the answer. If the answer doesn’t fit into the predefined criteria, the bot will notify the user and ask him/her to try again.
If Cersei would ask you a question, she would probably want to know your address so she could send her guards to fill your basement with barrels of wildfire.
Creating forms on Botsify
Templates
If you have problems building your first chatbot, templates can help you create the basic conversation structure. Unfortunately, not all platforms offer this feature for free. Snatchbot currently has the most comprehensive list of free templates. There you can choose a pre-built layout. The template selection ranges from simple FAQ bots to ones created for a specific industry, like banking, airline, healthcare, or e-commerce.
Choosing templates on Snatchbot
Plugins
Most tools also provide plugins that can be used for making the conversations more meaningful. These plugins allow Cersei to send images, audio and video files. She can unleash her creativity and make you suffer by sending you her favorite GoT execution videos.

With the help of integrations, Cersei can talk to you on Facebook Messenger, Telegram, WeChat, Slack, and many other communication apps. She can also sell her fan gear and ask you for donations by integrating in-bot payments from PayPal accounts. Her sales pitch will probably sound something like this:
“Gold wins wars! Would you rather invest your funds in a member of a respected family, who always pays her debts, or in the chaotic war endeavor of a crazy revolutionary, whose strength lies in three flying lizards? If your pockets are full of gold, you are already on my side. Now you can complete your checkout on PayPal.”
Chatbot building is now easier than ever, and even small businesses are starting to use the incredible benefits of artificial intelligence. If you still don’t believe that chatbots can replace customer service representatives, I suggest you try to develop a bot based on your favorite TV show, movie or book character and talk with him/her for a while. This way, you will be able to understand the concept that stands behind this amazing technology and use it to improve your business.
Now I’m off to talk to Cersei. Maybe she will feed me some Season 8 spoilers.
This article was originally published by Chatbots Magazine. Read the original post here.
Image credits for screenshots in post: Branislav Srdanovic
Banner stock media provided by new_vision_studio / Pond5 Continue reading

Posted in Human Robots