AI researchers need each advantage they can get when dealing with the unprecedented computational requirements of deep learning. GPU processing power has advanced aggressively, and chips at first designed to render images have become the workhorses powering world-changing AI research and development. Many of the linear algebra routines that are necessary to make Fortnite run at 120 frames per second are now powering the neural networks at the heart of cutting-edge applications of computer vision, automated speech recognition, and natural language processing.  

 

Now, the movement toward microchip specialization is changing into an arms race. Gartner projects that specialized chip sales for AI will double to around US $8 billion in 2019 and achieve more than $34 billion by 2023. Nvidia’s internal projections place the market for data center GPUs (which are almost solely used to power deep learning) at $50 billion in the same time frame. In the next five years, we are going to see significant investments in custom silicon come to fruition from Amazon, ARM, Apple, IBM, Intel, Google, Microsoft, Nvidia, Qualcomm. There are also a slew of startups in the mix. CrunchBase forecasts that AI chip companies, including Cerebras, Graphcore, Groq, Mythic AI, SambaNova Systems, and Wave Computing, have jointly raised more than $1 billion. 

 

To be clear, specialized AI chips are both important and welcomed, as they’re catalysts for transforming cutting-edge AI research into real-world applications. And yet, the flood of new AI chips, each one faster and more specialized than the next, will also seem like a throwback to the rise of enterprise software. We can expect cut-throat sales deals and software specialization focused at locking developers into working with just one vendor. Imagine if, 15 years ago, the cloud services AWS, Azure, Box, Dropbox, and GCP all came to market within 12 to 18 months. Their mission would have been to lock in as many businesses as possible — because once you are on one platform, it is tough to switch to another. This type of end-user gold rush is about to happen in AI, with tens of billions of dollars, and priceless research, at stake. 

 

Chipmakers won’t be short on promises, and the benefits will be real. But it is important for AI developers to understand that new chips that require new architectures could make their products slower to market — even with faster performance. In most cases, AI models are not going to be portable between different chip makers. Developers are well aware of the vendor lock-in risk posed by adopting higher-level cloud APIs, but in the past, the actual compute substrate has been standardized and homogeneous. This situation is going to change dramatically in the world of AI development.

 

It is rather possibly that more than half of the chip industry’s revenue will soon be driven by AI and deep learning applications. Just as software begets more software, AI begets more AI. We have seen it many times: Companies at first focus on one problem, but in the end solve many. For example, major automakers are striving to bring autonomous cars to the road, and their cutting-edge work in deep learning and computer vision is already having a cascading effect; the research is leading to such offshoot projects as Ford’s delivery robots. As specialized AI chips come to market, the current chip giants and major cloud companies will probably strike exclusive deals or acquire top performing startups. This trend will fragment the AI market rather than unifying it. All that AI developers can do now is understand what’s about to happen and plan how they’ll weigh the benefits of a faster chip with the costs of building on new architectures.

 

TRONSERVE

 

The median forecast was for 0.4%, according to a study by Nikkei Quick News. The slowdown was attributed to weak exports, which offset solid consumer spending and private investment. Japan had a 10-day holiday in May to mark Emperor Naruhito's ascension, which perhaps weighed on consumption and production, economists said.

 

The global economic outlook remains not certain in the face of heightened U.S.-China tensions. The International Monetary Fund in July lowered its overseas growth forecast for 2019 by 0.1 percentage point to 3.2%, although it said it expects growth to pick up to 3.5% in 2020. The Cabinet Office in July cut its forecast for Japanese growth to 0.9% from 1.3% for the year through March, 2020, quoting weak exports. That is equal to the latest IMF outlook.

 

Japan's exports fell, year on year, for the seventh straight month in June, as China's GDP growth slowed to 6.2% for April to June, the slowest pace since it began publishing data in 1992. Against this backdrop, Bank of Japan Gov. Haruhiko Kuroda said last week that he is 'more positive' about easing policy further to reach the central bank's 2% inflation target.

 

TRONSERVE

 

“We’re ready to start taking orders, and I already have a few,” Beckhoff said to a large group of end users and machine builders who raised hands quickly to be next in line. Attendees witnessed hovering magnets specifically carrying glass vials, packaged meat, and other consumer goods while flying with six degrees of freedom above a flat surface area with no rails, visible paths, or configurations.

 

“This intelligent transport technology does not replace conveyor belts,” said Uwe Prüssmeier, Beckhoff’s senior product manager, Drive Technology. “But XPlanar clears the way for completely new innovative plant concepts. It could completely change the way you build systems.”

 

The XPlanar principle is free floating, non-contact planar movers that hover over a surface area made up of either steel, glass, or plastic tiles — depending on the end user or OEM’s preference, industry, or application. Traveling magnetic fields generated in the planar tiles specifically move objects in any kind of pattern through various tracks, enabling batch one applications and flexibility in product handling. The movers are available in different formats to fit a range of applications, and the tiles move chiefly to accommodate production and the individual goods by automatically lifting, lowering, weighing, tilting, or rotating while traveling.

 

XPlanar is driven by Beckhoff’s TwinCAT 3 software platform for control and engineering and the company’s industrial PC scalable hardware platform. The contactless movers enable maximum freedom of movement, according to the company. Because the movers are controlled by magnetic fields, there are no mechanically moving parts, which eliminate the possibility of wear, vibration, and noise. The XPlanar can also be used in demanding environments since the surface area tiles can be hygienically-designed for food and pharmaceutical applications.

 

TRONSERVE

 

Advantech is using Nvidia’s Jetson platform in three AI edge systems to be produced later this year — the MIC-710IVA, MIC-720AI, and MIC-730AI. According to Advantech, these systems will grant AI application developers to rapidly create unique AI solutions based on Jetson. For cloud applications, Advantech’s SKY-6000 series servers with Nvidia’s graphics processing unit (GPU) enable the creation of high-density systems for big data aggregation. Featuring Advantech’s industrial design and Nvidia’s T4 GPU, these servers can resolve thermal issues mainly created by high-density GPU computing, according to Advantech.

 

These new announcements build on Advantech and Nvidia’s release earlier on this year of their AI Network Video Recording platform incorporating Advantech’s MIC-710IVA with Nvidia’s Jetson Nano. Nvidia describes Jetson Nano as a small computer on which different neural networks can be run in parallel for applications like image classification, object detection, segmentation, and speech processing. It can process using as little as 5 watts of power. “Advantech’s partnership with Nvidia is taking big steps forward in making AI a reality for manufacturing, transportation, and smart city applications. Through close collaboration in AI product development, Advantech and Nvidia are driving innovative transformations for AI applications,” said Linda Tsai, president of Advantech Industrial IoT Group.

 

TRONSERVE

 

The real hallmark of success for an automated self-service system is its capability to understand cases where an automated response is not appropriate. It needs to adequately recognize the need for human intervention, and seamlessly put the customer through to an agent where necessary. When situations escalate or get emotional, this is when swift access to a live agent is required.

 

What AI Requires to Do Well

 

An AI-based system learns the best way to handle situations based on data and experience. This means chatbots are only as good as both the data they are given to learn from and the process steps that have been pre-programmed to automate customer service requests. So many background data is required for an AI chatbot to be able to answer any given customer issue. There may be many hundred different ways that a customer may phrase a query, and several response options available to resolve the issue. For instance, a leading credit card company analyzed how many different ways customers asked for their present account balance. The outcome was 2,100 distinct ways—a surprising amount of variation for what seems a pretty prescriptive request at first glance.

 

It is true that machine learning allows AI to bit by bit improve performance over time. But at the outset, initial training is needed to customize the algorithms — the rules — for certain use cases.  Live chat and email logs can provide real-world data to capture customer intents.  Once these are evaluated and understood, organizations can then move towards a machine-learning chatbot implementation.

 

AI-Based Triage

 

When deciding where to apply chatbot technology, concentrating on processing only a small amount of very high-volume customer requests can often create bigger business wins. For instance, 50% of all customer inquiries may in fact relate to just three very common, regularly seen issues or faults.  Pointing chatbot learning and training to recognize and triage these inquiries properly will instantly improve the customer experience, while reducing agent workloads.

 

Natural language processing (NLP) in an integrated platform that balances human and AI-assisted service can discover whether an interaction would be better resolved by a human agent or AI. If human, it will route it to the best-skilled advisor; if AI, it will use Robotic Process Automation (RPA) to process the requests and answer to the customer.  NLP technology can also extend beyond a chatbot — playing an integral role in email customer service support or online chat, automating and accelerating response times.

 

Self-service technology is not here to replace or diminish communication with agents — it is as an alternative a tool to manage interactions more successfully. By judiciously applying intelligent routing with AI, those customers who need or expect agent intervention receive it flawlessly with a better experience — while those who have a routine query can rapidly access the information they need.

 

TRONSERVE

 

Sansiri will associate with local self-driving startup Airovr to implement transporting T77 residents to close by train stations by the end of the year.

 

Sansiri will also work with drone startup Fling to transport products from the community's malls to residences starting around March 2020. The services will be provided on a trial basis for six to nine months. Sansiri will come to a decision later whether to provide them full-time.

 

Established in 1984, Sansiri is renowned basically for high-end developments and also has partnerships with top Japanese developers. It plans to invest 600 million baht ($19.5 million) in real estate-related startups through a venture capital arm by the end of this year.

 

TRONSERVE

 

“This allows the Amazon Echo ecosystem to expand into most battery powered devices, which it couldn’t previously do,” says Syntiant CEO Kurt Busch. “You really couldn’t do always on listening with small battery powered devices, and now you can.” Sanjay Voleti, senior manager of device enablement with Alexa Voice Service was fascinated with Syntiant’s solution. “We’re excited to see developers begin using this technology in their devices and deliver new Alexa experiences for customers,” he said in a press release.

 

The chip attaches straight to a digital microphone or other sensor and initiates an “interrupt” line connecting to the larger—usually sleeping—system. Once that system’s yawned and expanded, it can then interrogate the NDP100 to determine what wake word or command it heard. The chip also keeps a three-second audio buffer if perhaps the system needs to catch up on what was said during its wakeup routine.

 

Like most AI ASICs, the chip only performs the inferencing step of deep learning. The training happens in the cloud using the common TensorFlow software library. Just one among the chip’s biggest advantages for developers, according to Busch, is that it doesn’t need any compiling or optimization step between the trained network and what goes on the chip. “It’s fair to say that the guts of our chip look like the Tensor Flow graph,” says Busch. Busch says Syntiant is now operating on a larger, more capable version of the chip to expand into more markets. It is also thriving a version that uses analog computation and embedded flash memory to do AI computations with less power.

 

TRONSERVE

 

The launching is likely to ramp up competition in the maturing global smartphone market as Chinese rival Huawei Technologies plans to release its own foldable phone the same month, while Apple is expected to introduce its latest iPhone in the fall. In addition to rectifying the faulty display, Samsung said it also improved its user experience as well, optimizing more apps and services for the new phone. The company emphasized the ability to multitask on the Galaxy Fold, which has a 7.3-inch display when unfolded. The display measures 4.6 inches folded, meaning it can be held in one hand.

 

'Now, with Galaxy Fold, you can review charts while talking on a video call, and search information, all at once,' said Samsung in a statement. 'Galaxy Fold completely reimagines the smartphone, changing the way you interact, communicate and experience everyday life.'

 

Just as Samsung, Huawei primarily delayed the launch of its own foldable phone, the Mate X. The Chinese company said it has carried out further testing of the device with carriers and developers and will begin selling it in September. The time of the two releases is not coincidental, according to analysts.

 

'It's interesting that Samsung and Huawei together again target September to officially make their foldable phones available,' said Joey Yen, an analyst at research company IDC. '[But] that's part of their hope, to have another round of marketing campaigns [at the same time that] this year's new iPhone, the iconic smartphone, although not a foldable one, will be introduced.'

 

Samsung was the world's largest smartphone maker by shipments in the first quarter with a 21% market share, as stated by Counterpoint Research, followed by Huawei with 17%. Apple came the third with 12%. Some analysts say Japan's export curb on South Korea could alter production of Galaxy Fold, as Japan's Sumitomo Chemical supplies a key display-making material to Samsung. The Japanese government imposed the restrictions earlier in the month, which require domestic suppliers to obtain approval from the authorities to export hydrofluoric acid, photo resists and polyimides to South Korea.

 

Hydrofluoric acid, also known as etching gas, and photo resists are essential components in producing chips, while polyimides are used for manufacturing display panels. 'An end-product that could be affected is the Samsung Galaxy Fold smartphone. The display of the Samsung Galaxy Fold is produced utilizing fluorinated polyimide film from Sumitomo Chemical, which is a Japanese electronic materials firm,' said Tadashi Uno, display research director at IHS Markit. But Uno said that Samsung has a plan to replace Sumitomo. 'South Korea-based Kolon Industries could act as an alternative supplier for the Samsung foldable smartphone display.'

 

TRONSERVE