Google Announces New Cloud TPU VMs For AI Workloads

By

-

June 30, 2021

Google announced their new cloud TPU VMs that will authorise companies to supply instances with Tensor Processing Units (TPU). Virtual machines (VM) will offer enterprises a new and advanced user experience to develop and deploy TensorFlow, PyTorch, and JAX on Cloud TPUs.

The Cloud TPUs need not be accessed remotely over the network, rather the virtual machines can be set up in an interactive development environment on each TPU host. For example, now machine learning models can be debugged line by line using a single TPU VM and later be scaled up on a superfast Cloud TPU Pod slice to take advantage of the TPU interconnect.

The users will have access to every TPU VM created by them at all times and can run any code in a tight loop with TPU accelerators. In addition, local storage can be used to execute custom code in input pipelines and easy integration of Cloud TPUs into production workflows.

The Cloud TPU VMs are available in two variants — Cloud TPU v2, which is based on the second-generation TPU chipsets, and the new Cloud TPU v3 based on third-generation TPU. According to Google Cloud, the significant difference between both is in the performance factor. While the Cloud TPU v2 can perform up to 180 teraflops, the TPU v3 up to 420 teraflops. Both variants can be used to perform tasks such as artificial intelligence-powered healthcare analytics and quantum chemistry.

The new TPU system is built in a much simpler and flexible way to achieve performance gains. As the code need not make round trips in the data center network to reach the TPUs the overall efficiency is improved. It is also cost-efficient as the data processing can run directly on the Cloud TPU hosts and eliminates the need for additional Compute Engine VMs.

Atos Launched An Artificial Intelligence Powered Platform ‘ThinkAI’

By

Dipayan Mitra

-

June 30, 2021

Atos Launched An Artificial Intelligence Powered Platform ‘ThinkAI’

Atos has recently launched its new end-to-end scalable artificial intelligence-powered platform ThinkAI. This unique platform will enable organizations to successfully design, develop, and deliver high-performance artificial intelligence software.

The platform is dedicated to the companies that want to run more accurate and faster simulations using traditional high-performance computing. ThinkAI would allow enterprises to process and analyze large amounts of complex data in a very short time by integrating artificial intelligence and high-performance computing.

This technology of Atos will also help companies to reduce their carbon footprints to create a competitive advantage. Alex Norton, data analysis manager at Hyperion, said, “The Atos ThinkAI solution brings together the important pieces for HPC users, at all levels of their artificial intelligence journey, to leverage the significant opportunities of AI in their own research.”

He further explained that this solution could help customers to combine the latest hardware and software technologies to speed up the development of critical artificial intelligence-based solutions. ThinkAI is the most comprehensive high performance computing and artificial intelligence solution available in the market.

It is a wholesome platform that provides consultation and hardware-software solutions and delivers rapid results at reduced operational costs.

The ThinkAI solution framework consists of –

‘Advise’ with industry-contextualized consulting supported by experts at The Atos Center of Excellence.
‘Architect’ using best in class artificial intelligence hardware and software, including partnerships with Graphcore and NVIDIA, aided with Atos’ digital security.
‘Accomplish’ with end-to-end solution orchestration, accelerating time to artificial intelligence operationalization and industrialization at a reduced cost.

Agnes Boudot, Senior Vice President of Atos, said, “Artificial intelligence has created a new paradigm for applications in the scientific and industrial domains, catalyzing the processing of data to actionable insights. As the complexity of machine learning models and its associated costs continue to grow substantially, dedicated high-performance artificial intelligence infrastructures become crucial for organizations that want to deliver research breakthroughs. “

Samsung Launched Artificial Intelligence-Powered Vacuum Cleaner

By

Dipayan Mitra

-

June 29, 2021

Samsung Launched Artificial Intelligence-Powered Vacuum Cleaner

Samsung has recently launched its artificial intelligence-powered robot vacuum cleaner named Jet Bot AI+, which was first revealed at CES 2021. The initial launch has allowed deliveries in the United States and Europe.

Samsungs plans to launch the product globally by the latter half of 2021. The robot vacuum cleaner comes with a unique active 3D sensor capable of accurately scanning a wide area to prevent any head-on collisions with smaller objects.

The sensor is capable of detecting objects as small as one centimeter. Jet Bot AI+ can trap more than 99.99% dust, including 0.5~4.2µm sized microparticles, and can be easily controlled via voice command. It uses Intel’s artificial intelligence solutions to improve its object recognition technology further.

Hyesoon Yang, Vice President and head of the customer experience team of Samsung, said, “Jet Bot AI+ reflects Samsung’s ongoing dedication to building connected appliances that deliver dynamic performance that can adapt to the varying needs of our customers.”

He added that they are very excited about introducing this robot with the world’s best recognition technology. This will now enable users to clean their homes intelligently and efficiently using artificial intelligence-powered Jet Bot.

The robot vacuum cleaner is also equipped with a LiDAR Sensor that calculates its location precisely to optimize its cleaning path by continuously scanning the area to gather distance information. It has the ability to work seamlessly even in dark or poorly lit areas with a negligible number of blind spots.

Consumers can also monitor their home remotely with the Jet Bot AI+’s Patrol Mode. The unit’s front camera livestreams video images through the SmartThings app.

The mode allows consumers to monitor the status of rooms they want to check while away. Privacy related issues of the patrol mode have been taken care of, as it encrypts the video clips so that only authorized users can view them.

DataRobot 7.1 Introduced Enchantments To Enhance Artificial Intelligence

By

Manisha Kalagara

-

June 29, 2021

DataRobot released version 7.1 with several upgrades to its Augmented Intelligence platform to level up artificial intelligence projects. The enhancements were built on the enchantments that were announced at DataRobot’s AI Experience Worldwide last month.

DataRobot 7.1 introduced new MLOps Management Agents that provide remote lifecycle management for enterprises. Management Agents automate various tasks and determine any remote model’s state despite their fabrication and current status. With the snowflake push-down integration, both DataRobot and Snowflake developers will access automatic discovery and new feature computations for their models directly in the Snowflake Data Cloud.

With Automated Time Series, DataRobot runs its Eureqa forecasting models as a part of the regular Autopilot process. Using the intelligent feature selection, Eureqa models exponentially reduce complications and function better with large datasets. To enhance the artificial intelligence experience, the new version introduced a new No-Code AI App Builder that allows developers to quickly develop artificial intelligence models and help with decision making and what-if scenarios.

“We are in constant communication with our customers regarding the challenges they face when deploying AI, and as a result, will tailor our updates based on their unique needs,” said Nenshad Bardoliwalla, SVP of Product at DataRobot. He also added that DataRobot is committed to building a platform that empowers every individual and encourages non-technical business users to take advantage of artificial intelligence.

Version 7.1 includes automated data prep for time series, nowcasting for time-aware models, automated AI reports, job prediction, and scheduling UI to manage and maintain prediction schedules. The detailed features can be checked from the website.

Triton Plans To Create 24,000 Jobs In Telangana For New EV Factory

By

Dipayan Mitra

-

June 29, 2021

Triton plans to create 24,000 jobs in telangana

US-based electric vehicle manufacturer Triton recently announced that it had signed an agreement with the government of Telangana to set up an electric vehicle manufacturing facility in Zahirabad.

The contract was signed by Jayesh Ranjan, Principal Secretary of IT and Industries, and Triton Electric Vehicle Pvt. Ltd CEO & Founder Himanshu B. Patel in the presence of the chief minister of Telangana.

The enterprise said that the new facility would create more than 24,000 jobs in the state. In a recent statement, Triton EV said that it intends to invest $1.5 billion to build and improve its electric vehicle manufacturing facility in the next five years.

Himanshu Patel said that they are planning to expand rapidly to meet the high surging demand for electric vehicles in the country. For this, the company had looked for investment opportunities in various states of India before selecting Telangana because of its industry friendly policies.

Tesla’s rival Triton to create 24,000 jobs in Telangana with new electric vehicle factory https://t.co/fALyqEMUV3 via @BiIndia
— KTR (@KTRTRS) June 24, 2021

Kalvakuntala Taraka Rama Rao, Minister for Municipal Administration & Urban Development, Industries & Commerce, and IT of Telangana said, “We are very excited to know about Triton EV’s expertise and vision for the electric vehicle market. Industrial policy of the state which is dedicatedly crafted towards fast process driven decision making and action towards bringing new investments in the state.”

The company plans to penetrate not only Indian but other markets like Sri Lanka, Bangladesh, Nepal, and middle eastern countries with its manufacturing facility in Telangana. Tesla, the company’s biggest rival, recently announced that it is also ready to enter the Indian market by setting up a new research and development center in Bengaluru.

Triton EV’s officials confirmed that their products include zero-emission electric vehicles, electric passenger cars, semi-trucks, buses, and three-wheelers. These vehicles are part of the company’s integrated transportation ecosystem which also includes charging infrastructure and other digital products.

Artificial Intelligence Powered Digital Assistant Is Modernizing Latin America’s Mining Industry

By

Dipayan Mitra

-

June 29, 2021

According to recent research, an artificial intelligence powered digital assistant is modernizing Latin America’s mining industry. The assistant has been developed by a Chilean startup InDiMin. The unique platform provides real time feedback on processes, predicts events, and suggests possible improvement opportunities at mine sites.

Co-founder of InDiMin, Loreto Acevedo said, “I believe Latin America has incredible human resource, mining knowledge and a unique diversity, which means that it also has the capability to foster real, radical change.” She further added that her enterprise’s product has played a major role in the modernization of mine sites like Chile’s Los Colorados iron mine.

The first prototype of this platform was released in 2013 which was developed by Acevedo and her partner Alvaro Diaz. The duo interviewed a number of people at medium and large mine sites to understand their requirements and to find out what more can be done with technology to further modernize the industry.

Acevedo said in an interview, “In these past years, we’ve pioneered artificial intelligence developments for the mining value chain, from planning to drill and blast, load and haul, mineral recovery and grinding processes.” She also mentioned that they have achieved 25% improvement in the platform’s performance which has allowed big mining companies to gain more than $7.5 million in Latin America.

Back in 2012, Loreto launched a project to incorporate semi-autonomous drill jumbos at a mine site. But due to the reluctance of the workers to accept the technology, the project was forced to shut down.

Loreto Acevedo firmly believes that the way things are done within the mining industry can be revolutionized by proper implementation of artificial intelligence powered technologies and sustainable practices.

Amazon Announces AWS BugBust Challenge To Fix Software Bugs

By

Manisha Kalagara

-

June 29, 2021

Amazon announced the AWS BugBust challenge, which is a global competition that challenges developers to eliminate software bugs. Any software developers from around the world can join the challenge by registering on the AWS amazon website.

As the world of coding has taken over almost every sector, the volumes of codes have gone from thousands of lines to millions of lines for better operations.

But bugs are almost unavoidable while developing any code. A large amount of time and effort is taken to eliminate these bugs from the huge codebase. The AWS BugBust is the largest bug bash challenge, aiming to improve code and costing quality using the power of machine learning to eradicate 1 million bugs.

Organizations can nominate an administrator to set up the AWS BugBust event in the AWS console. The administrator can invite participants in a few clicks and set up the challenge. Once the challenge is set up, the Amazon CodeGuru scans the code and finds all the bugs for the team to start busting. Once the bugs are solved, the team member will receive points depending on the number of bugs busted. These points will be added to the organization’s leaderboard and the global leaderboard to compete with developers globally.

There are many points, prizes, and rewards, from t-shirts to trophies, as you go up the leaderboard. The top 10 bug busters globally will receive an expense-paid trip to attend AWS re: Invent in Las Vegas, where they can participate in the live BugBust challenge.

The AWS BugBust will change the face while helping enterprises save money and also by making codes more efficient. Technologies like machine learning and automated reasoning are assisting developers to be better, by powering tools like Amazon CodeGuru.

Max Healthcare Launched AI Enabled Remote Patient Monitoring Network

By

Manisha Kalagara

-

June 29, 2021

Max Healthcare has launched an artificial intelligence-powered remote patient monitoring network in association with MyHealthcare. The device-integrated patient monitoring framework will help patients across India connect to the Max Healthcare doctors from the comfort of their homes.

Patients can monitor their vitals with clinical devices connected to the Max MyHealth app. The app monitors these readings and forwards them to the electronic medical records (EMR) for the doctor’s review. It can monitor Spo2 levels, blood pressure, blood sugar, ECG, and heart rate.

All these vitals are tracked via artificial intelligence tools that aid with ECG interpretation and send alerts to the doctors in the nearest healthcare center if the parameters range outside permissible limits. The launch of the app during the second wave of Covid-19 will help with the crowding of hospitals, reducing the spread of the virus by eliminating trips to the hospitals.

“At Max Healthcare, our constant endeavor has been to use the advancements in digital technology to provide best in class medical support to our patients. Our focus is to expand the geography of care for the Max Healthcare group,” said Prashant Singh, Director IT and Group CIO, Max Healthcare.

He also mentioned that the launch of the remote patient monitoring platform is an initiative to help and enhance the healthcare services to patients at their homes. The service will help extend post-hospitalization services to Tier 2 and Tier 3 cities, making quality healthcare accessible for a wider population.

In the press meet, Max Healthcare announced that patients would get access to various complex care programs involving daily/regular inspection such as cardiac care, diabetes, and hypertension management and monthly/bi-weekly virtual consultations with Max Hospital doctors.

Self-Driving Startup Pony.ai Considers Going Public

By

Dipayan Mitra

-

June 29, 2021

Self-Driving Startup Pony.ai Considers Going Public

Self-driving startup Pony.ai, backed by Toyota Motor Corp, is now considering going public in the United States to help fund its motive of commercializing driverless ride-hailing services.

The company is currently active in the United States and China, and it plans to install its technology in hundreds of vehicles by 2022, rising to tens of thousands in 2024-2025, said the chief executive of Pony.ai.

Recently, the company announced that it aims to launch its Robotaxis in the United States by 2022.

CEO of Pony.ai, James Peng, said to Reuters, “For autonomous driving, it’s a big opportunity. But at the same time, it’s also a long term, big opportunity.” He further mentioned that in order to achieve their goal, the company would require to raise more funds to support their research and operations.

The company, on Friday, said in a statement that it had tapped the vice chairman of investment banking at JPMorgan Chase & Co, Lawrence Steyn, as their chief financial officer to help the company grow its customer base at a global scale.

Many other Self-driving startups, including Pony.ai, like Alphabet Inc’s Waymo and General Motors Co’s Cruise, have also been racing to raise funds as the industry prepares to scale up operations. Peng said to Reuters that they are still considering and debating about the time frame for a public share sale.

He also said that this fresh move by the startup is yet another way of raising funds. According to Peng, the next big challenge for the self-driving vehicle industry is to reduce the manufacturing cost while speeding up their expansion to different parts of the world.

The company was founded by former employees of Google and Baidu in 2016. Pony.ai has raised over $1 billion, including a $462 million funding from automobile manufacturing giant Toyota.

IIIT Hyderabad Announces Online Course On Foundations Of Machine Learning

By

Dipayan Mitra

-

June 28, 2021

IIIT Hyderabad Announces Online Course On Foundations Of Machine Learning

IIIT Hyderabad recently announced the launch of its new 50-week foundation course in Modern Machine Learning for engineering students.

The online course is available for undergraduate engineering students who are in their third or fourth year pursuing a program in Information Technology, Computer Science, Electrical Engineering, Electronics Engineering, or other related branches from any technical institution in India.

Nominations of participants have to be placed by universities which IIIT will scrutinize before declaring the list of selected students. This 50-weeks certificate program in Modern Machine Learning will offer online modules with a personalized learning experience for each student.

The course will also pay equal attention to the foundation and practical implementation of machine learning. It would also provide discussion opportunities with industry leading researchers from academics.

Core instructors of this online course include eminent professors of IIIT Hyderabad, including Anoop Namboodiri, Ravi Kiran Sarvadevabhatla, and C. V. Jawahar.

The participation fee for the course is INR 10,000 + tax, with an opportunity for deserving students to get financial assistance. The course will also provide cash rewards of INR 25,000 to the top 5% of participants after completion. INR 15,000 to the following top 20%, INR 10,000 to the next top 25%, and INR 8000 to all the students who will complete the course, respectively.

Prof C V Jawahar, Dean research and development, said, “Since IIITH has the largest artificial intelligence and machine learning academic lab in the country we’ve started this course to allow learners from various colleges to gain from our expertise and set them on their career paths should they choose to specialize in machine learning in the future.”