The startups' ecosystem in India with a focus on AI/ML and emerging tech is growing exponentially. According to research we did recently - despite the pandemic and the resulting lockdowns, the AI startups in India have attracted total funding of $836.3 million — the largest funding outlay in the last seven years at a 9.7% year-on-year growth — last year.
Our Startups segment each week focuses on these startups. Last Sunday, we covered Voiro, the Bangalore-based startup leveraging emerging technologies to bring automation and intelligence to leading media organisations across the world.
New Delhi-based Astra Security provides an end-to-end security suite for threat protection, malware monitoring, vulnerability assessments and penetration tests.
We got in touch with Vidit Baxi, Co-founder of Safe Security, to talk about the challenges in cyber risk management, the use of AI & ML by cybercriminals, the future of cyber risk management, and more.
We got in touch with the founders of Locobuzz — Vishal Agarwal (co-founder and CEO), Shubhi Agarwal (co-founder and COO), Nitin Agarwal (co-founder and CTO) — to get insights into their startup story, use of AI and ML in their offerings, growth plans and more.
Is Mainframes A Dying Technology?
Introduced in the 1950s, Mainframes are high-performance computers with large memory and processing power that can handle billions of calculations and transactions in real-time. They are superior to PCs, work stations, and minicomputers in more than one way.
Today, every enterprise and industry are moving to the cloud. So, the question is, does a technology introduced more than half a century ago still hold water?
AI Safety In a Complex World
Researchers at DeepMind surveyed six use cases where an RL agent behaviour can be monitored. In their recently published technical report, the researchers have presented a few methodologies that can be used to predict failure in AI systems.
Their experiments were carried out through the lens of cause-effect relationships that can explain why an agent behaves in a certain way.
Yann LeCun and his team now proposed a new method called Barlow Twins. Named after neuroscientist H.Barlow, this method draws heavily from his influential 1960 article, titled ‘Possible Principles Underlying the Transformation of Sensory Messages’.
Barlow Twins method applies redundancy reduction, similar to Barlow’s one in his article, to self-supervised learning. Barlow Twins is conceptually simple, easy to implement and learns useful representations.
Featured Video |Do Large Machine Learning Models Struggle At Maths?
The New US-India Artificial Intelligence Initiative
The Indo-US Science and Technology Forum launched the US India Artificial Intelligence Initiative on March 17 to foster the science and technology relationship between the two countries.
Analytics India Magazine spoke with geopolitics experts to understand what to expect from the initiative and how India will benefit from the partnership
Hands-On Guides for ML Developers
Python 3.10: What To Expect
Python started work on the upcoming version, Python 3.10 pre-alpha, in May 2020. Python version 3.10 is set to be released on 4 October 2021. However, with the pre-alpha, the company has unveiled an early developer preview of Python 3.10.
Transformers For Vision
Transformers are all geared up to rule the world of computer vision. The runaway success of OpenAI’s CLIP and DALL.E had a lot to do with the sudden interest in multi-model machine learning in research circles.
Apart from the recent OpenAI releases, other works have also attracted a great deal of attention in this subdomain.
BOTTOM OF THE NEWS
Here's what all happened last week.