- 5 deep learning startups to follow in 2017
- MacBook Pro review: Lovable despite the short battery life
- Google’s AI assistant has 5 New Year’s resolutions for you
- My New Year’s resolution is to delete Twitter from my phone
- AWS sees growth in database migrations
- Chinese firms reportedly ordered to pay Disney, Pixar $194,000 for copying ‘Cars’
- HTC: No Vive 2 at CES
- Spanish social advertising company Adsmurai raises $4.2 million
- Do chatbots really help you stay productive?
Posted: 01 Jan 2017 12:10 PM PST
If artificial intelligence (AI) hadn’t hit the mainstream before, it did this year. Google chief executive Sundar Pichai came through with the best sound bite, saying that the world is going from being “mobile-first” to “AI-first.”
Apple squished AI into the iPhone, and Google stuck it in the Pixel. Facebook brought it to the News Feed, and Microsoft put it in Word. Samsung bought AI startup Viv to catch up with Apple’s Siri virtual assistant. And messaging apps like Skype and Messenger now feature chatbots.
Much of the AI attention has been focused on deep learning, which entails training artificial neural networks on lots of data and then getting them to make inferences about new data. And in the past five years, more and more deep learning startups have cropped up.
This year, chipmaker Intel bought Nervana, which was making hardware and software for deep learning, and enterprise software company Salesforce bought MetaMind, which made deep learning software that could quickly process lots of images and text. Both Nervana and MetaMind appeared on my list of five deep learning startups to follow in 2015. Meanwhile, the startups on my 2016 list are all chugging along.
Now I have a new batch of companies to watch for in the year ahead:
Bay Labs is among the startups applying deep learning to medical imaging. Its engineering-heavy team includes Johan Mathe, who previously worked on Google’s Project Loon. Yann LeCun, director of Facebook’s artificial intelligence research group, has invested in the startup, along with Khosla Ventures.
Cerebras is a secretive startup headed up by Andrew Feldman, who sold microserver company SeaMicro to AMD for $334 million. Feldman’s new startup is building AI hardware, and esteemed venture capital firm Benchmark led a funding round totaling more than $20 million, according to a source familiar with the matter. (Feldman wouldn’t comment for this article.)
Based in Palo Alto, Deep Vision is building low-power chips for deep learning. While at Stanford, two of the startup’s cofounders, Rehan Hameed and Wajahat Qadeer, cowrote an interesting paper laying out a “Convolution Engine chip multiprocessor.”
Graphcore has developed an intelligent processing unit (IPU) PCIe accelerator that neural networks can use for training and making inferences. The startup has also built software for working with its infrastructure using the existing MXNet and TensorFlow deep learning frameworks. Investors include Bosch Venture Capital, Foundation Capital, and Samsung Catalyst Fund.
Founded in 2012, ViSenze performed better than several groups in certain parts of the 2016 ImageNet image recognition competition. Funded by Rakuten Ventures, ViSenze is a spinoff of NExT, a research center established by National University of Singapore and Tsinghua University of China. Its software can perform object recognition and tagging for images and videos and provide visually similar content.
Posted: 01 Jan 2017 09:05 AM PST
The 2016 MacBook Pro is scandalous. Relative to its predecessor, it’s missing ports, short on battery life, skimpy in the keyboard department, and more expensive.
Me, personally? I really like this laptop, even with its imperfections.
It is thinner and lighter. More of the case is made out of aluminum. The speakers are louder. The trackpad is larger. And if you pay extra, there’s a Touch Bar display above the keyboard that gives you a new way to control things. And it comes in space gray!
I will admit that, feature by feature, a 2015 MacBook Pro may be a better choice for some. (HDMI port? SD card slot? USB-A? MagSafe? Yeah.) But the new version is well polished. Do I wish I could upgrade it myself? Of course. Could the battery last longer? Oh, yes.
The thing is, it’s the finest laptop I’ve ever used.
Apple has come up with an excellent answer to Microsoft’s Windows Hello feature in Windows 10, which lets people sign in by doing nothing more than looking at a PC’s display — specifically infrared cameras. Apple didn’t have to invent anything new; it simply added the Touch ID fingerprint scanner that’s been available on iOS devices for three years. But instead of doubling as a home button, it serves as a power button on the MacBook Pro models that come with the Touch Bar.
It’s not as fast as fingerprint scanners on other devices, like Huawei’s Mate 9 phone. And you can’t always use Touch ID to authenticate — like when you restart, for example. But when it is available as an option, it speeds up the process of signing in and buying things on desktop.
The Touch Bar
Right up with Touch ID, the Touch Bar is the main attraction of the 2016 MacBook Pro. The biggest question I’ve asked myself about the Touch Bar since Apple first revealed it is whether it’s something you must have. After using the base-model 13-inch MacBook Pro with Touch Bar as my main computer for a month and a half, I’ve concluded that you can live without it. In fact, Apple advises developers against exposing features exclusively in the Touch Bar.
That said, the Touch Bar can provide an easier way to access features that are ordinarily hard to find. That’s true with the Ulysses writing app, for example.
And I don’t miss the F row of keys that the Touch Bar replaces as much as I thought I would. The software escape button at the left end of the Touch Bar usually does the trick. And I like that Siri now has its own dedicated button.
Could Apple make the Touch Bar brighter? Yes. Could more app developers optimize for it? Absolutely. I still like it. I look forward to seeing it on Apple’s standalone keyboards and less expensive MacBooks.
The 13.3-inch, 2,560-by-1,600px Retina display is damn clear. It’s more precise than any laptop I’ve tested. It makes additional connected monitors less interesting. It gets quite bright, too, at 500 nits. Watching high-definition video and zooming in on quality photos is a treat.
Keyboard and trackpad
The keyboard does indeed resemble the extremely shallow one on the 12-inch MacBook Pro. The Pro is thankfully more comfortable to type on, though; you don’t need to push the keys so hard to make your keystrokes known. I’ve grown to love writing on it for long stretches of time — even at night it’s fine because the keyboard backlight lights up every key evenly, unlike HP’s Spectre x360. But during quiet hours I try to tap more gently, because the keys can make a good amount of noise.
The trackpad is definitely giant. And I like it that way; I haven’t encountered problems with palm rejection when I’m typing.
The overall look and feel of the laptop is top-choice as well. The edges of the aluminum case are satisfyingly sharp. It still feels sturdy even though it is a tenth of an inch thinner and half a pound lighter than the previous generation. Nothing here feels flimsy.
Generally speaking, this computer can handle heavy computational loads. It’s certainly more powerful than the 12-inch MacBook. But it does occasionally stumble. Under the weight of a few dozen browser tabs, for example, it can slow down.
The laptop I’ve been testing has a 2.9GHz sixth-generation Intel Skylake Core i5 chip, an Intel Iris Graphics 550 integrated graphics processing unit (GPU), 8GB of LPDDR3 RAM, and a 256GB solid-state drive (SSD). Given the $1,800 price, one might hope for more.
The same goes for the battery. Apple advertises up to 10 hours of wireless web usage on all variations of the new MacBook Pro, but I’ve never gotten more than around eight hours on the 13-inch Touch Bar model, with its 49.2 watt-hour battery. My average was below five hours. Put another way, I get less out of one charge than I did on the 2016 12-inch MacBook Pro — and my early 2014 13-inch MacBook Air.
Yes, you can live with low battery life if you bring an external battery everywhere you go. I’ve come to think of this as a necessity for the current MacBook Pro. It’s not convenient — it does cancel out the weight savings! — but it’s workable.
Leave your charger at the office? No problem — if you have a USB-C charger handy. I’ve got one here from HP at the moment — you’ll be fine. It just works. This is one upside of the arrival of USB-C on the Mac.
Having four USB-C ports, up from one on the 12-inch MacBook, is freeing. It’s proven to be enough for me. But the important thing here is what’s not present. I’ve gotten by with just a couple of USB-A to USB-C adapters and a VGA Multiport adapter. People with more complex needs may have to rely on more dongles, which can be tiresome.
All in all, I feel very productive on the new MacBook Pro. Its usual responsiveness is engaging, and typing on it feels natural.
But it could be more powerful given its price tag. So people will need to decide for themselves if they should pay more to get improved specs on the Touch Bar models. (The non-Touch Bar version may be out of the question for some because it has just two USB-C ports.)
Either way, in the coming years Apple will inevitably enhance this laptop just like it does all of its products. When that happens — when it gets new chips and longer battery life, and when there are thousands of apps that integrate with the Touch Bar — the MacBook Pro will be a better choice.
Posted: 01 Jan 2017 08:35 AM PST
Google Assistant and the smart speaker Google Home have some opinions on how to be a better person in 2017.
Ask “What should my New Year’s resolution be?” and the AI assistant will tell you to do things like write a novel or pick up calligraphy.
Here are the five answers to the question Google Assistant gave VentureBeat when we asked earlier today.
Nothing earth-shattering or controversial about this ability, but it’s not a question Alexa or Siri attempt to answer.
Google Home also has an opinion on the best way to lose weight, one of the most common New Year’s resolutions. When asked “How do I lose weight?,” Google Home serves up a top result, one of those results that sit in a grey box at the top of a search query.
Google Assistant didn’t have much to say about other common New Year’s resolutions. The assistant was unable to answer questions like “How do I exercise more?” or “What’s the best way to save money?”
As simple or gimmicky as the ability to dole out New Year’s resolution advice may be, a distinct personality may have an impact on sales and adoption, as assistants like Cortana from Microsoft, Siri from Apple, and Alexa from Amazon attempt to distinguish themselves.
Intelligent assistants are fighting for space in mass-market devices, not just smart speakers or smartphones.
For example, in recent months, plans have been announced to integrate Alexa into a lamp, vacuum, and watch.
Google Home first became available in October for $129. Expanded actions like the ability to order pizza or get news briefings that are now available on Google Home will be made available for Pixel smartphones and Allo chat app in 2017.
Posted: 01 Jan 2017 06:28 AM PST
I can’t do this anymore.
I don’t enjoy staring blankly into this abyss every night, my clenched jaw lit by the glow of someone else’s stream of consciousness (not even in chronological order).
Scanning my feed like an addict doesn’t help me do my job — though once I swore it did. I can’t cash in my follower chips for a good story or pay my bills with the weak dopamine reward of a few faves and RTs. I don’t need to plow every fleeting observation and passing thought into Twitter’s advertising machine. I don’t have to workshop jokes in real time.
The nagging urge to fill my spare seconds with nonsense and sweet owns doesn’t widen the boundaries of my world. Twitter doesn’t make me any smarter.
My responsibilities, goals, and happiness in life extend beyond the walls of Twitter. So for now I’m deleting the app from my phone.
Phew, it’s gone.
Posted: 01 Jan 2017 03:34 AM PST
Public cloud market leader Amazon Web Services (AWS) offers a lot of services — 83 by my count. Some of them are more popular than others. Without question, the core EC2 computing service and S3 storage service are among the most popular. But then what?
AWS’ parent company, Amazon.com, doesn’t break out sales numbers for all of the AWS services. So instead AWS spectators have to lean on occasionally released, typically non-financial figures to get a feel for what’s being used.
On Friday AWS chief executive Andy Jassy shared new data on increasing use of the Database Migration Service (DMS), which lets organizations quickly and easily move databases from on-premises data centers onto AWS’ data center infrastructure. AWS first introduced it at its re:Invent conference in October 2015.
“DB [database] freedom is a powerful thing,” he wrote.
The numbers — which follow Amazon’s April 28 disclosure that in the year to date “more than 2,000 databases” had been migrated with DMS — show a clear spike, and they don’t appear to be directly attributable to any one enhancement announced at re:Invent 2016.
But Jassy’s comment about “freedom” is a bit more complex. With the help of the free Schema Conversion Tool, customers can use the Database Migration Service to take data from existing databases — SQL Server, Oracle, PostgreSQL — that cost lots of money and push their data into cloud-hosted versions. While AWS offers cloud implementations of all of those, AWS also runs its own Aurora managed relational database.
While adopting Aurora might mean freedom from Oracle and Microsoft databases, customers can simultaneously become shackled to Aurora, because, although it’s advertised as being 1/10 the price of “commercial databases,” it’s not open source, even if it is compatible with MySQL and PostgreSQL.
In any event, leaving aside the “freedom” bit, what we do know today is that usage of the Database Migration Service is increasing.
Posted: 31 Dec 2016 10:14 PM PST
(Reuters) — A Shanghai court ordered two Chinese firms to pay Walt Disney Co and Pixar more than 1.35 million yuan ($194,440) compensation for copying parts of their hit movies “Cars” and “Cars 2”, the official Xinhua news agency reported on Saturday.
The ruling is the latest in a slew of intellectual property wins for large foreign firms, who have complained about widespread copyright infringement in China.
Disney and Pixar took the Chinese firms to court saying the characters, titles and posters from local animation “The Autobots” were substantially similar to those from “Cars” and “Cars 2”.
The court agreed that the Autobots characters K1 and K2 were similar to Disney and Pixar’s animated cars Lightning McQueen and Francesco Bernoulli, Xinhua said.
The court ordered infringement activity to stop immediately, and said Disney and Pixar should receive 1 million yuan to cover economic losses, as well as 350,000 yuan for legal expenses.
Disney is making a major push into China with the recent opening of a $5.5 billion theme park in Shanghai, its first on the mainland. Its animated movies including “Zootopia” and “Big Hero 6” have been big box office hits there.
Disney, Pixar and the two Chinese firms were not immediately available for comment.
Xinhua said the total order covered more then 1.35 million yuan, but did not list any other payments.
German carmaker BMW and basketball star Michael Jordan have both won intellectual property cases in China this year.
(Reporting by Engen Tham)
Posted: 31 Dec 2016 08:15 PM PST
I'd wager most people who bought the HTC Vive love the unit but wish a new version would bring key improvements. A slimmer design and lighter cord, a better fit for the face and more ergonomic controllers without hard-to-reach grip buttons are all near the top of the early adopter wish list.
HTC relies on Valve's lighthouse tracking technology to make the Vive and its controllers work. Earlier this year, Valve opened up that technology to a wide range of partners. While none of these partners have officially announced products compatible with this SteamVR Tracking technology yet, early buyers are anxious because future headsets and accessories could be lower cost if you already have the VR tracking system installed.
This means anticipation is high for CES next week. Really high. The event, spread across Las Vegas, is one of the world's biggest consumer electronics conferences. Last year, HTC used CES to offer a first hands-on demo with the Vive Pre developer kits, giving early adopters their first look at what consumer room-scale VR looks like. Given this anticipation, rumors have been circling that HTC is set to reveal the next iteration of its Vive VR headset at CES. To put it plainly: they're not true.
HTC confirmed in a note to VR news site UploadVR that it will not be announcing a "Vive 2" at this year's show.
"There is no truth to the rumor of launching Vive 2 at CES 2017," the official statement from HTC reads. "At Vive, we are laser focused on building out a strong and growing ecosystem for current and future Vive owners so they can experience the best room-scale VR with the most compelling content available."
It is also worth noting that, while we expect major announcements and VR-related reveals next week, Oculus is not listed as an exhibitor at CES this year. Facebook is, but there is no booth listed. Last year at CES Oculus had a very large booth on the show floor at the Las Vegas Convention Center. That doesn't appear to be the case this time around. It makes sense on multiple levels, given that Facebook seems to be absorbing the Oculus brand, and they just shipped Oculus Touch a couple of weeks ago. CES isn't really necessary for Facebook because if you want to check out their latest VR efforts you can go into one of hundreds of stores around the country for a demo of the Oculus Rift.
This story originally appeared on Uploadvr.com. Copyright 2016
Posted: 31 Dec 2016 07:03 PM PST
Barcelona-based social advertising company Adsmurai has received €4 million ($4.2 million) in a second round of funding led by venture capital firm Axon Partners Group, with participation from Banc Sabadell, through its program BStartup10, and Enisa, a Spanish government-funded financing group.
Launched in 2014 by Marc Elena, Otto Wüst and Juan Antonio Robles, Adsmurai specializes in generating and optimizing advertising campaigns on social networks. A partner of Facebook, Instagram, YouTube and Twitter, the company manages its clients' ad campaigns on these platforms for an approximate value of €20 million.
Adsmurai is headquartered in Barcelona and has offices in Madrid, Mexico, Colombia and Peru.
The company intends to put the funds toward the development and improvement of their services using artificial intelligence applied to image recognition, as well as to promote their international expansion, with a focus on Europe.
Read more: WebCapitalRiesgo
You can subscribe to Tech.eu’s newsletter here.
This story originally appeared on Tech.eu. Copyright 2016
Sign up for Funding Daily: Get the latest news in your inbox every weekday.
Posted: 31 Dec 2016 04:41 PM PST
When Slack burst onto the workplace, employees rejoiced. Finally, there was a way to chat without having to send a dreaded email or worse, get up and actually go talk with your coworkers face-to-face. Thanks to Slack and a handful of other messaging platforms, business teams could easily communicate using a single interface that would allow them to also send files, GIFs, and more.
The role of chatbots
Workplace messaging platforms like chat typically come with artificial assistants, or chatbots. Think of the Slackbot or iPhone's Siri or Amazon's Echo. These pieces of software can chat with employees and are also used to assist companies with everyday tasks. Siri and Echo can relay the time, weather, and even order an Uber for someone on the go with a simple command and internet connectivity. These chatbots, of all shapes and sizes, utilize a process called deep learning, which mimics neurons in the neocortex and learns to recognize patterns in digital representations of sounds and other data.
Chatbots provide convenience, streamlining certain processes that would cost time and energy for human employees to execute. For example, chatbots can help companies train new employees by moving entire training sessions online. These artificial intelligence platforms can also deliver data-driven results in places like call centers, assisting customer service representatives as they solve problems for frustrated customers. Chatbots could even replace your human assistants in some situations — instead of asking the front desk to book your flights or appointments, your chatbot assistant could do that for you with a simple text command.
But when it comes to workplace culture and productivity, how much are these artificial intelligence platforms helping your company?
As with any new technology rollout, chatbots come with some initial problems that could hinder workplace productivity. Chatbots and artificial intelligence programs need to be able to understand the specific request a user asks, even if the request doesn't come in the form of a question. Turing Tests have exposed a weakness in artificial intelligence, revealing how difficult it will be to build more natural-sounding chatbots and to train computers to interpret the written word. For example, a comprehensive chatbot assistant must be able to discern colloquial language, slang, and formal language, as well as separating out multi-step requests that could come in a single sentence.
If you implement a chatbot into your workplace, consider the following scenario. On the one hand, a chatbot could help you accelerate your research for a specific project, or even help you find documents hidden deep within your folders. But what happens to productivity when employees need to stop to clearly spell out an inquiry because the chatbot misunderstood the first time? Having to fix a chatbot's mistake could actually set your workers back and waste their time. Similarly, using an HR chatbot to vet potential employees can be problematic, as a person's resume consists of more than just their technical skills and experience. AI software would need to be able to understand an applicant's soft skills and be able to make the connection between their work experience and a company's HR handbook.
The good news for intelligent chatbots is that technology is advancing at a rapid pace, and bots are learning how to process the human language better with each passing day. And there seems to be a growing acceptance of chatbots and an increased understanding of the potential benefit they offer.
Convenience and productivity
Facebook's Mark Zuckerberg built an AI assistant, just like Iron Man's own assistant Jarvis, to control his home and manage his work projects at the same time. Zuckerberg and his team are also building an AI to answer questions directly on Facebook Messenger. When applied to Facebook Workplace, this could dramatically increase employee productivity. While most bots can only answer simple questions at the moment, the idea is to one day build a platform that can maximize a worker's productivity by passing mundane tasks off to an artificial assistant.
Despite current hesitations about chatbots, large enterprise software companies are investing money into developing "smartbots" to process some employee tasks. What many developers anticipate is that businesses will use bots to replace slow, outdated websites that make it difficult for them to effectively communicate with their customers. In particular, chatbots and AI platforms provide businesses with an opportunity to link all aspects of their company in a single layer that can be managed by a bot. This means an employee can use text-based commands to tell a chatbot to update aspects of a company's website, instead of having to code it themselves. It's an opportunity to improve productivity that will run into a few setbacks initially but could have dramatic implications for how we conduct business in the future.
Bots versus humans
On the flip side, we have to ask how chatbots will impact workplace culture. The rise of machines inevitably brings up the question of human capital and whether certain jobs will become extinct once artificial intelligence and chatbots can assume those roles. Certainly, there is an economic benefit that comes with using chatbots. Businesses can save money and resources by "hiring" robots to replace humans who would require a salary, benefits, and more. As deep learning becomes more complex and machines are taught to think like the human brain, the case could be made that specific jobs, like the role of an assistant or data analyst, could go extinct. Productivity would likely also increase, as machines could process requests faster than people can, but this raises the question of what will happen to those who find themselves out of a job. And with fewer people working in offices and interacting with one another, it's worth wondering what company culture will look like in the future.
At the end of the day, how a business chooses to implement new technologies will determine whether a chatbot improves or hinders workplace productivity. Without a doubt, businesses can benefit from a little extra help, and employees are free to focus on important issues when chatbots have taken over the most mundane tasks. Certainly, proper training and an improved deep learning software can streamline your office productivity. As for how bots will influence your culture? That remains to be seen.
|You are subscribed to email updates from VentureBeat. |
To stop receiving these emails, you may unsubscribe now.
|Email delivery powered by Google|
|Google Inc., 1600 Amphitheatre Parkway, Mountain View, CA 94043, United States|