Connect with us


Google launches service to automate machine learning



Courtesy Google
Share this:

Google has announced a new service designed to help democratise AI. The company is targeting businesses which want to access machine learning but don’t have the resources to hire an AI team. Its cloud can now auto-generate machine learning models.

Cloud AutoML Vision

Called Cloud AutoML, the set of services gives businesses automated AI building tools. They can then create machine learning-infused services without a detailed understanding of the underlying tech. Cloud AutoML consumes data supplied by the business to automatically generate a machine learning model. This is then optimised to suit the dataset and begin producing further results.

The initial release of the platform will include one service, Cloud AutoML Vision. This provides a drag-and-drop interface to create new machine learning models using image recognition. Users only need to upload a set of images to start training and optimising an AI model that recognises them. Once the model is confirmed to be working, it can be deployed from AutoML straight to Google Cloud.

Google sees AutoML as a significant development of its AI product portfolio. The company recognised that one of the biggest limiting factors for AI is the lack of experts skilled in the subject. Although demand from businesses is high, there’s evidence that AI deployments are being restricted by the shortfall in developers.

Google’s positioning AutoML to help address this growing problem. Businesses that want to use AI in common scenarios, such as image recognition, could use Google’s self-service automated solution instead of their own in-house team.

Faster, cheaper AI solutions

This could dramatically cut costs and increase the turnaround time when creating new machine learning models. Google said AutoML will move the industry forward by allowing less-skilled developers to start creating complex AI applications.

“Currently only a handful of businesses in the world have access to the talent and budgets needed to fully appreciate the advancements of ML and AI. There’s a very limited number of people that can create advanced machine learning models,” said Google. “To close this gap, and to make AI accessible to every business, we’re introducing Cloud AutoML. We believe Cloud AutoML will make AI experts even more productive, advance new fields in AI and help less-skilled engineers build powerful AI systems they previously only dreamed of.”

Cloud AutoML is the result of a cross-company development effort at Google involving collaboration between the company’s Cloud, Brain and other AI teams. Google said development of AutoML is ongoing and several new products for the platform are being prepared for release.

Businesses interested in the initial computer vision solution can register their interest now. Availability is initially restricted to a select number of users but should be broadened to more Google Cloud customers over time.

Share this:


AI cybersecurity provider Vectra raises $36m in funding



Share this:

Vectra has raised $36m in Series D funding in an investment round led by Atlantic Bridge. The cybersecurity company has developed an AI which uses machine learning to monitor corporate networks. Admins get advance warnings if potential problems are found.

Announced in a press release today, the new investment round takes Vectra’s total funding to $123 million. Besides the Atlantic Bridge technology equity fund, the Ireland Strategic Investment Fund, Nissho Electronics Corp and several previous Vectra investors also participated in the latest round.

Bringing AI to the cybersecurity fight

Vectra’s AI proactively detects anomalous network traffic before it has a chance to gain a foothold. The company is one of several providers that’s aiming to automate cybersecurity to improve accuracy and resilience. Security admins get warnings as soon as a problem is found, without having to spend time manually analysing server logs or records of network traffic.

The startup already has over 400 enterprise customers, with 60 percent of them sharing data back to the company. This information is used to tune Vectra’s AI so it’s more capable of finding potential entrypoints and identifying emerging risks. Vectra told VentureBeat that customer data sharing helps the company’s AI to better protect all of its users by sharing insights through the cloud.

Real-time cyber threat detection

Vectra will use the $36 million in new investment to expand its business and reach new enterprises. It currently employs over 140 people in various countries worldwide. The company will make additional hires to grow its expertise in artificial intelligence. The additional talent will help make Vectra’s AI more effective at pinpointing cyber threats in real-time.

“We have been impressed by the remarkable growth of Vectra in this fast-moving cybersecurity market,” said Kevin Dillon, managing partner of Atlantic Bridge. “The increasing volume, creativity and effectiveness of cyberattacks means that enterprises must adopt AI to automate cybersecurity operations. We look forward to helping the company expand its global enterprise footprint.” Vectra expects to add over 200 new people to its workforce across technical development, AI research and sales and marketing. The company intends to accelerate the growth of its platform and business to reach new enterprises across the world.

Vectra is competing with a growing roster of companies using AI to thwart cyberattacks, including major cloud providers such as IBM and Microsoft. The company is succeeding by focusing on “non-stop” automated hunting of threats, using behavioral models that find and isolate attackers before they’ve breached the frontline defences.

Share this:
Continue Reading


Canadian Cloud region brings efficiency, agility and AI to businesses



cloud computing
Share this:

Canada’s first Google Cloud region is open for business in Montreal. The new region, called “northamerica-northeast1”, is the 15th for Google worldwide, and the fifth in North America

With its Canadian offering Google joins Amazon Web Services (AWS), which has run a Canada-based region since 2016.

The new region could boost Google Cloud use, and cloud infrastructure in general, for Canadian businesses that have yet to make the switch.

Global Head of Solutions for Google Cloud Miles Ward said the new region will improve latency for Google Cloud’s Canadian customers.

“To be able to put this infrastructure into Montreal improves not just the performance in Montreal, but for all customers here in Canada,” said Ward.

Ottawa-based company Pythian provides consultations and services for international companies, and CEO of Pythian, Paul Vallée, spoke of the advantages that the new cloud region brings to companies looking to build their North American presence.

“You don’t need to put in us-east1 anymore, you can put it right here (in Montreal) and still have access to the entire low latency dynamics, and to the entire North American economy which is I think something new and something really exciting.”

Making the switch to cloud

Google’s new region adds to Canada’s ongoing innovation in the cloud infrastructure arena. According to Forbes, Canada is a major player in the push towards public cloud strategies. The federal government recently released its plan to move all unclassified data to the cloud.

But some hesitation remains. Many business owners will be weighing whether the benefits outweigh the costs and effort to transition to cloud options.

For Ward, the public cloud is all about agility, efficiency and remaining competitive in an innovative market.

“The reality is that the efficiency of these centralized resources is orders of magnitude higher. As a result, the companies that are able to take advantage of those tools just are more agile — able to make choices more quickly with lower risk, able to operate at lower cost. The result is this is the opportunity of this generation to leapfrog their competitors, to outcompete and to operate on a global stage. “

Vallée stressed that the speed of cloud-based project work is the key advantage for businesses — and that this new method of storing and sharing work outmaneuvers old ways of defining success.

“A lot of companies need to adopt cloud because of a velocity or a business agility imperative. They’re adopting the technology in order to win by beating their competitors to market — not win by saving pennies, and not win through a more efficient capital structure, but win because you beat them there and you built it before they could.”

Google Cloud offers modern, AI-friendly option to businesses

Having a data centre located in Montreal could also sway businesses with concerns about data sovereignty or latency to embrace Google Cloud.

According to Vallée, the new region is just the latest reason businesses will want to get on board with Google for their public cloud services, over competitors like AWS or Microsoft Azure.

“They have differentiated in two major ways versus the other cloud vendors,” said Vallée. “The first one is they have a very simplified, platform-as-service-oriented cloud. That’s their roots, that’s their DNA. The other cloud platforms really started with infrastructure-as-a-service and started tacking on platform features after the fact.

“Whereas Google went the opposite route. Their Cloud started with Google apps for MyDomain and… they’ve been expanding that so that now they’re roughly comparable (to other vendors), but have very much a platform centre, which is a much more modern approach to cloud infrastructure.”

Vallée also sees a big bonus to companies looking to develop AI-related products and tools within Google Cloud.

“The other dynamic that I think is really important is Google is really a leader in machine learning, and has the most compelling demonstrations of their machine learning capability in terms of their road map. Google is, I think by general concensus, far in the lead in terms of their machine learning innovation, what with their Google Brain project and Google DeepMind intiative.

“And what we’re seeing coming out of Google Brain and Google DeepMind is all being built into the Google Cloud API support over time. Which means that for future proofing your cloud investment, if you are doing an analytics-oriented, data science-oriented, machine learning or AI-oriented adoption of the public cloud, Google is really a nicely differentiated platform to make that kind of investment on.”

Share this:
Continue Reading


Canada uses blockchain to make research grants more transparent



Share this:

The Canadian government has begun trialling blockchain technology as a way to improve the transparency of research grants. The National Research Council (NRC) is currently using an Ethereum-based system to publish funding information in real-time.

In a blog post, the NRC explained how blockchain technology could help to make government contracts more transparent. The blockchain’s public ledger means data recorded on the system is unalterable and open to everyone. This provides transparency into the workings of government, which in turn promotes trust.

Public-private partnership to drive pilot

To implement the trial project, the NRC has partnered with Canadian blockchain SME Bitaccess. It’s also working with the Industrial Research Assistant Program (IRAP), a body that generates a large volume of transactions each year and which would benefit from improved transparency.

Using funding from the Build in Canada Innovation Program, the NRC and Bitaccess are piloting a blockchain record-keeping system for the IRAP’s financial activities.

The program is part of the Canadian government’s wider efforts to improve transparency and utilize modern technologies. The NRC will be responsible for investigating how the blockchain could be applied to other areas of government. If the pilot proves successful, Canada could begin using blockchain more broadly to preserve public records and maintain transparency.

The trial is described as the first “real-use case” of its kind for deploying blockchain tech inside public institutions. The NRC said it expects to acquire “constructive” insights into how blockchain could be used by government bodies. Many tech visionaries see blockchain as crucial to the future of business but it’s still a new concept to most official organisations.

“These are early days yet, but the experiment is expected to provide constructive insight into the potential for blockchain technology and how it may be used for more open and transparent function of public programs,” said the NRC. “This experiment also marks an important step forward for the technology and a commitment by the Government to support emerging Canadian innovation.”

From cryptos to conservation

The blockchain is currently best known as the infrastructure supporting cryptocurrencies such as Bitcoin. In this scenario, the blockchain records every transaction on a decentralised public ledger. As the blockchain is immutable and distributed across computers around the world, the data stored within is always secured against external tampering.

These qualities are also what makes the blockchain concept attractive to organisations that need to store data transparently. The Canadian government’s initiative is just one example of how the tech could be used. Other recent blockchain-based projects have included schemes aimed at sports fans and unsustainable practices in the tuna industry.

Share this:
Continue Reading


Herding cats, weekly

* indicates required