Monthly Archives: June 2017

In 1950, Alan Turing, already famous for helping to crack the German Enigma code during World War II, devised the Turing test to define intelligence in machines. Could a computer, Turing asked, fool a human into thinking he was interacting with another person, or imitate human responses so well that it would be impossible for a person to tell the difference? If the machine could, Turing proposed, it could be considered intelligent. Turings thought experiment spawned scores of science-fiction tales, such as the 2015 hit movie Ex Machina. Now, artificial intelligence (AI) and autonomous algorithms are not only passing the Turing test every day but, more importantly, are making and saving money for the businesses that deploy them.

CenturyLink is one of the largest telecommunications providers in the United States, serving both small and large businesses nationwide. The company collects thousands of sales leads from the businesses it serves, and it wishes to interact with them in the intimate, personal manner consumers have come to expect. Pursuing those leads more effectively would accelerate the companys growth, and converting and upselling a larger percentage of hot leads (people who have expressed interest in the companys services by filling out a form, clicking on an ad, or emailing the company) would boost the companys bottom line.

Accordingly, in the latter half of 2016, CenturyLink made a small investment in an AI-powered sales assistant made by Conversica to see if it could help the company identify hot leads without hiring an expensive army of sales reps to comb through the leads. The Conversica AI, a virtual assistant named Angie, sends about 30,000 emails a month and interprets the responses to determine who is a hot lead. She sets the appointment for the appropriate salesperson and seamlessly hands off the conversation to the human.

The potential customer gets a prompt and helpful outreach from Angie, and the reps who may each have 300 accounts save time because Angie vets the inquiries to identify the ones with the most potential. The reps also become more efficient because Angie routes the right leads to the right reps. In the small pilot CenturyLink ran, Angie could understand 99% of the emails she received; the 1% that she couldnt understand were sent to her manager.

According to Scott Berns, CenturyLinks Director of Marketing Operations, the company has approximately 1,600 sales people, and the Angie pilot started with four of them. That number soon rose to 20, and continues to grow today. Initially, Angie was identifying about 25 hot leads per week. That has now increased to 40, and the results have certainly validated the companys investment. It has earned $20 in new contracts for every dollar it spent on the system.

Tom Wentworth, Chief Marketing Officer at RapidMiner, a company that provides an analytical tool for data scientists, had a problem that was similar to CenturyLinks. Like many software companies, RapidMiner offers free trials, and Wentworth was struggling to serve the approximately 60,000 users who come to the companys site each month for the free trial. Many of the visitors using RapidMiners software, and needing help, are not paying anything for the service. So, how could Wentworth help them in a cost-effective way?

The company had a popular chat feature on its site, but its salesforce was overwhelmed and spending a great deal of time sorting through the chat sessions to find potential customers. It was like looking for the proverbial needle in a haystack.

Wentworth approached a friend who suggested he try a chat tool called Drift, which would ask a visitor initiating a chat, What brought you to RapidMiner today? The visitor would respond, and the Drift bot would provide one of seven potential follow-up answers. For example, a visitor might say, I need help, and Drift would send him or her to the support section of the website.

Drift was relatively easy to set up. Wentworth, like CenturyLink, started small, running the tool on a few of RapidMiners smaller web pages to test how helpful it was.

In less than two weeks, he had deployed it on every page.

The Drift bot now conducts about a thousand chats per month. It resolves about two-thirds of customer inquiries; those that it cannot, it routes to humans. In addition to Wentworth, who is monitoring the tools interactions, two co-op college students support the inquiries part-time. Wentworth told me that Drift is generating qualified leads for the sales team by making customers. Its the most productive thing Im doing in marketing, he said.

Every day, Wentworth reviews conversations people have had with Drift. Ive learned things about my visitors that no other analytics system would show, said Wentworth. Weve learned about new use cases, and weve learned about product problems.

This is the strength of an AI agent that can elicit information like a person, rather than an analytics tool that simply finds patterns in the data it collects, like a machine.

In 2016, Epson America, the printer and imaging giant, piloted the same Conversica AI assistant as CenturyLink. Chris Nickel, Epsons senior manager of commercial marketing, was drowning in all the leads he was getting for the companys diverse line of products: big printers, projectors, scanners, point of sale solutions, and industrial robots. Epson America was getting 40,000 to 60,000 leads per year from trade shows, direct mail, email marketing, social media, print and online advertising, and a successful brand awareness campaign. The leads would pour in, and whether they were good, bad, qualified or not, they would all be turned over to salespeople whose availability to follow up was inconsistent.

After implementing the AI assistant, Epsons leads are now followed up promptly and persistently until their AI assistant gets a response. Because the outreach to leads takes 6-8 times, Conversica is a true force multiplier for our sales team, say Nickel. After a lead is passed to one of Epsons partners, the AI assistant follows up to make sure the customer was satisfied. Sometimes, the response to that follow-up identifies a new sales opportunity, such as everything went great, and actually we are looking to buy another 60 projectors, giving Epson the opportunity to quickly capitalize on a new sales opportunity before the competition. Or it can uncover an unresolved customer support issue, such as Im having a problem with my projector.

As Nickel told me, Before, if we gave 100 leads to the reps, we might get a couple of responses from customers. Now, if we give 100 leads to the AI assistant, we get 50 responses. Epson reports that the official response rate with the AI assistant is 51%, representing a 240% increase from the baseline established at the beginning of the pilot, and a 75% increase in qualified leads. According to Nickel, that has produced $2 million in incremental revenue in just 90 days.

Because the AI tools that Epson America, RapidMiner, and CenturyLink deployed are offered as-a-service, it was easy for these companies to conduct pilots, and then scale up. Clearly, its worthwhile for companies to test AI-powered chat or email tools to see if they can convert more leads, and improve their understanding of what customers want and need.

When it comes to AI in business, a machine doesnt have to fool people; it doesnt have to pass the Turing test; it just needs to help them and thereby help the businesses that deploy them. And that test has already been passed. As one CMO told me, AI tools are the only way I can scale helpfulness to a global community of 200,000-plus users with a team of two.

View post:

How AI Is Streamlining Marketing and Sales - Harvard Business Review

Size isnt everything, relationships are

imageBROKER/REX/Shutterstock

By Matt Reynolds

The world is a confusing place, especially for an AI. But a neural network developed by UK artificial intelligence firm DeepMind that gives computers the ability to understand how different objects are related to each other could help bring it into focus.

Humans use this type of inference called relational reasoning all the time, whether we are choosing the best bunch of bananas at the supermarket or piecing together evidence from a crime scene. The ability to transfer abstract relations such as whether something is to the left of another or bigger than it from one domain to another gives us a powerful mental toolset with which to understand the world. It is a fundamental part of our intelligence says Sam Gershman, a computational neuroscientist at Harvard University.

Whats intuitive for humans is very difficult for machines to grasp, however. It is one thing for an AI to learn how to perform a specific task, such as recognising what is in an image. But transferring know-how learned via image recognition to textual analysis or any other reasoning task is a big challenge. Machines capable of such versatility will be one step closer to general intelligence, the kind of smarts that lets humans excel at many different activities.

DeepMind has built a neural network that specialises in this kind of abstract reasoning and can be plugged into other neural nets to give them a relational-reasoning power-up. The researchers trained the AI using images depicting three-dimensional shapes of different sizes and colours. It analysed pairs of objects in the images and tried to work out the relationship between them.

The team then asked it questions such as What size is the cylinder that is left of the brown metal thing that is left of the big sphere? The system answered these questions correctly 95.5 per cent of the time slightly better than humans. To demonstrate its versatility, the relational reasoning part of the AI then had to answer questions about a set of very short stories, answering correctly 95 per cent of the time.

Still, any practical applications of the system are still a long way off, says Adam Santoro at DeepMind, who led the study. It could initially be useful for computer vision, however. You can imagine an application that automatically describes what is happening in a particular image, or even video for a visually impaired person, he says.

Outperforming humans at a niche task is also not that surprising, says Gershman. We are still a very long way from machines that can make sense of the messiness of the real world. Santoro agrees. DeepMinds AI has made a start by understanding differences in size, colour and shape but theres more to relational reasoning than that. There is a lot of work needed to solve richer real-world data sets, says Santoro.

Read more: The road to artificial intelligence: A case of data over theory

Read more: Im in shock! How an AI beat the worlds best human at Go

More on these topics:

Read the original here:

DeepMind's neural network teaches AI to reason about the world - New Scientist

Over the past three months, Nvidia's (NASDAQ:NVDA) stock has been upgraded by several financial services firms including Goldman Sachs (NYSE:GS), Citigroup (NYSE:C), and Bernstein, while some others have downgraded the stock, such as Pacific Crest. In an article published in December, last year, I said Nvidia's stock could scale new highs if the company's revenue continues to grow at a CAGR of 20% plus in the foreseeable future. At that time, the stock created a new high around $120, before correcting almost 20% afterwards.

I also cautioned investors that the stock could go through spine-chilling volatility, and that's exactly what is happening now. The commentaries of several sell-side analyst firms fueled the extent of volatility beyond what happens under normal circumstances. My medium-term target for the stock is $200+, albeit with continued volatility. The new catalyst for the stock will be the availability of its HGX-1 platform at the hyperscale space.

Nvidia: Revisiting the Bull Thesis

My investment thesis was based on the expectation that Nvidia's revenue will grow at a CAGR of 20% plus in the next three years. The primary revenue growth driver will be the expanding application of artificial intelligence (AI) and/or deep learning (DL) across various industries. Let's focus on Nvidia's competitive advantage in AI/DL.

Nvidia's DGX-1 platform is its trump card to lead the next generation of AI-wave leveraging the TensorFlow software library, originally developed by Google (NASDAQ:GOOGL)(NASDAQ:GOOG). Nvidia recently launched a partner program with hyperscale vendors, such as Foxconn, Inventec, Quanta and Wistron, for its HGX-1 AI reference platform. What's the difference between DGX-1 and HGX-1? Well, HGX-1 is the hyperscale version of the DGX-1 platform. The launch of the partnership program will have a strong and sustainable impact in the HPC (high performance computing) space via hyperscale players.

Investors need to understand how the advantages of TensorFlow coupled with Nvidia's HGX-1 reference architecture will boost its revenue in the foreseeable future, which Nvidia's stock at today's price around $150 hasn't factored in.

CUDA, HGX and TensorFlow

TensorFlow has become extremely popular since its inception in late 2015 by the Google Brain Team. Its appeal lies in its usage of data flow graphs for solving complex mathematical problems related to building efficient neural networks. Add to that its support for multiple GPUs and CPUs via a single API. This is the secret sauce behind the immense success of TensorFlow, out-competing other AI-related software libraries, such as torch, caffe and theano in terms of speeding up the training process of the neural networks.

Engineers at Nvidia were quick to understand TensorFlow's competitive advantage over other software libraries. While Intel (NASDAQ:INTC) was busy to make its MKL (math kernel library) more versatile via incorporating the Neon deep learning framework of Nervana, the startup which Intel acquired almost a year ago, Nvidia silently made its CUDA parallel computing software platform compatible with TensorFlow.

The DGX-1 platform, which is basically a parallel processing-based hardware accelerator with eight Nvidia Tesla P100 GPUs, is the platform that gave birth to HGX-1. Both DGX-1 and HGX-1 connect the eight Tesla P100 GPUs via NVLink interconnect. With the introduction of HGX-1 earlier this year and then making it available to end-users via different hyperscale vendors, Nvidia has opened the door for running different kinds of AI/DL workloads for various industries.

How HGX-1 would expand the scope of DGX-1 via taking the latter to the hyperscale space? Okay, before addressing the issue let's delve a bit deeper into how Nvidia's upcoming Tesla V100, i.e., Tesla powered by Volta GPU architecture, will help DGX-1 deliver much faster performance compared to other GPU-based systems. The Volta architecture will be made to support Tensor-based AI/DL operations, which implies the HGX-1 will also deliver Tensor-based performance boost at the hyperscale space.

HGX At Hyperscale: What Does It Really Mean?

What does HGX at the hyperscale space mean for investors? Simple. It means more revenue for Nvidia. Since hyperscale computing enables datacenters to deploy distributed computing environments across the globe via scaling only a few servers to lots of servers, Nvidia's GPUs will see a steady rise in adoption. However, that doesn't necessarily mean all the datacenters across the world will need to deploy the high end Tesla P100 or the upcoming Tesla V100 GPUs.

Only the hyperscale datacenters will need them. For the traditional datacenters, Nvidia's Pascal-powered midrange Quadro workstation GPUs will be enough. This essentially means HGX has the ability to boost the overall demand for Nvidia's server-grade GPUs. Nvidia's decision to launch the HGX partner program will help it grab a significant portion of the AI/DL market share from archrivals Intel and AMD (NASDAQ:AMD).

Intel is also trying to make MKL compatible with TensorFlow. However, the process isn't complete yet. I believe Intel has potential to compete with Nvidia in the parallel processing space. But due to its FPGA-dominated business model and late entrance in the AI/DL marketplace, it won't be able to become a real competitor of Nvidia just yet.

Unfortunately for Nvidia, though, AMD has made significant progress in the last six months. AMD, with its upcoming Radeon Instinct line of GPUs along with its upcoming Zen-based Naples CPUs (bundled with the SenseMI technology), could be Nvidia's real competitor, but for that to happen the OpenCL (open computing language) software library (which is the CUDA equivalent for AMD, and to some extent Intel) needs to support TensorFlow. Please refer to my article mentioned at the beginning to learn more about CUDA and OpenCL.

Valuation

Nvidia's trailing 12-month revenue is $7.5 billion. A CAGR growth rate of 20%, which is quite possible due its AI/DL leadership as analyzed above, will catapult its revenue closer to $13 billion in 2020. And since such an incredible growth is almost given, I won't be surprised if the Street offers a P/S multiple of 17/18x to its stock in the next six months. After a year, Nvidia's revenue will be around $9 billion at the CAGR rate mentioned above, and at ~15x P/S multiple the stock will get closer to or even slightly beyond the $200 level. So it's a strong possibility that the stock will cross the $200 mark in the next six to twelve months.

NVDA Revenue (TTM) data by YCharts

Final Words

Nvidia is a secular bull story. However, you need nerves of steel to be a part of the story. AI/DL was a concept even a few years ago, and has started to take the shape of an industry just now. I would recommend risk-tolerant investors to hold the stock as a long-term investment.

Disclosure: I am/we are long NVDA, GOOGL.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Link:

Nvidia's Next Big Thing: The HGX-1 AI Platform - Seeking Alpha

Mashable

An AI server took China's national math exam and did pretty badly Mashable An AI machine that sat the math paper for China's college entrance exam has failed to prove it's better than its human competition. AI-Maths, a machine made of 11 servers, three years in the making, joined almost 10 million high schoolers last week, in ... Even An AI Supercomputer Found This College Entrance Exam ToughIFLScience all 6 news articles »

View original post here:

An AI server took China's national math exam and did pretty badly - Mashable

A US intelligence director says a lot of espionage is more boring than you might think, and much of it could be handed over to artificial intelligence.

A significant chunk of the time, I will send [my employees] to a dark room to look at TV monitors to do national security essential work, Robert Cardillo, head of the National Geospatial-Intelligence Agency told reporters including Foreign Policy. But boy is it inefficient.

Cardillo calls out recent advances in artificial intelligence, giving algorithms the ability to analyze vast amounts of images and video to find patterns, give data about the landscape, and identify unusual objects. This kind of work is critical for assessing national security concerns like foreign missile-silo activity, or even just to check in on North Korean volleyball games.

Cardillo has hired a former tech CEO, Anthony Vinci, to lead development of this machine-learning technology. Vinci previously founded Findyr, a company that crowdsources data for companies, like pictures of how products are displayed on shelves or infrastructure development progress.

But the US government is already trailing behind Silicon Valley on this pursuit. Facebook, deep into its crusade to connect the world, was able to apply machine learning to satellite data last year, analyzing buildings likely to contain wireless internet down to five-meter accuracy. Those data were intended to be used to guide Facebooks internet drone, Aquila. (Aviation might prove a more difficult project than machine learning: Facebooks head of global aviation policy indicated that the company doesnt have a timeline to even get one drone in service after last years test.)

Google has also touted its ability to discern details in satellite imagery for accuracy-critical uses, like defense and aviation. Stanford University has used satellite imagery to map poverty.

Cardillos initiatives arent the first use of AI by intelligence or defense agencies. DARPA and IARPA, US defense and intelligence research agencies, have been funding AI research for decades, and the Central Intelligence Agencys venture arm is supporting efforts to apply AI analysis to satellite imagery.

Read more here:

US intelligence agencies are beginning to build AI spies - Quartz

Artificial intelligence may be all the craze in Silicon Valley, but on Wall Street, well, theres a lot of skepticism.

High-powered algorithms are not a new phenomenon in finance, and for this industry, the name of the game is efficiency and precision.

Quite frankly, finance executives want systems that, in one way or another, make money. Because of this, new wild and flashy AI systems that just make something smart wont fly.

The fintech companies that are successfully leveraging AI today are the ones that have found a very concrete way to apply the technology to an existing business problem. For example, technology such as specialized hardware, big data analytics, and machine learning algorithms are being used in fintech to augment tasks that people already perform.

At the Singularity University Exponential Finance Summit this week, Neil Jacobstein, faculty chair of Artificial Intelligence and Robotics at SU, shared some of the most interesting AI companies in fintech right now.

Not surprisingly, these companies each have a clear market application and reduce friction in the business problems they address.

Numerai is a new kind of hedge fund that is built by crowdsourcing knowledge through a massive network of hedge fundsthe system collects hundreds of thousands of financial models and individual predictions. With this information, Numerai is building their own financial models that incorporate the algorithms submitted through the crowdsourced community. Numerai has already secured funding from First Round Capital and Union Square Ventures, which is no small feat.

In 2010, AlphaSense launched its intelligent search engine, which uses AI, natural language processing algorithms, and advanced linguistic search tools to provide researchers with critical insights with serious accuracy and speed. Financial analysts can pose questions to AlphaSenses systems and get insights that are significantly more customized and accurate than a simple Google search would provide. Its a great example of an AI augmenting a critical task in finance: research.

Opera is helping companies turn their big data into predictive insights and business intelligence. The company uses pattern recognition to identify what they call signals, meaning actionable insights from data. Their signals help researchers understand conditions that may be happening in the market, or the world at large, so they can act quickly on these changes.

AppZen is a very practical solution to one of every executives most arduous taskssubmitting expense reports. The system uses AI to audit 100 percent of employee expenses and then generates an expense report in real time. Automating this process saves companies hours of lost productivity. AppZen also gives companies more confidence in their ability to flag suspicious charges. So, if youve been considering expensing that pricey night out with clients, dont, because AppZen will likely flag it.

CollectAI is a cloud-based software system thats shaking up the collection business. The system is able to mimic the voice and tone of a collection agent to gather important information over the phone about a collections case. With this information, CollectAI uses a self-learning algorithm to learn about the case, and then pulls knowledge from previous successful cases and applies those insights to decide how to best approach the situation at hand. The system gets better and better over time, which is pretty incredible.

Image Credit: Pond5

Go here to read the rest:

5 Fintech Companies Using AI to Improve Business - Singularity Hub

CAMBRIDGE, MA: For all the speculation about where the outer reaches of artificial intelligence may lie, brands can usefully deploy AI today in order to both make and save money.

In an article for the Harvard Business Review site, consultant Brad Power outlined how AI-powered tools are already making a significant difference to businesses as diverse as telecoms providers, software companies and imaging businesses.

"AI tools are the only way I can scale 'helpfulness' to a global community of 200,000-plus users with a team of two," one CMO explained.

Virtual assistants are able to send emails and manage responses, to conduct chats with customers and handle routine inquiries, to identify potential leads and pass the best ones on to humans.

For a business like RapidMiner, whose free trials of an analytical tool for data scientists were attracting 60,000 users a month, chat tools offered a way of managing far more interactions than its existing staff could ever do.

According to chief marketing officer Tom Wentworth, its chosen tool resolves around two-thirds of customer inquiries while also generating sales leads. "It's the most productive thing I'm doing in marketing," he told Power.

"I've learned things about my visitors that no other analytics system would show," he added. "We've learned about new use cases, and we've learned about product problems."

Power suggested that the ability of an AI agent to elicit information like a person rather than simply looking for patterns in data is a particular strength.

And because many of these tools are now offered as-a-service, it is simple for businesses to carry out trials and scale up.

"Clearly, it's worthwhile for companies to test AI-powered chat or email tools to see if they can convert more leads, and improve their understanding of what customers want and need," Power said.

Evidence of just how worthwhile it can be came from Epson America, the printer and imaging business: a 75% increase in qualified leads produced a claimed $2m in incremental revenue in just 90 days.

Data sourced from Harvard Business Review; additional content by WARC staff

Here is the original post:

AI scales 'helpfulness' - Warc