Tuesday, January 22, 2019

NPTEL Online Course on Deep Learning

A total of 118 Videos created by NPTEL on Deep Learning. Visit the below URL for the entire playlist:
https://www.youtube.com/playlist?list=PLyqSpQzTE6M9gCgajvQbc68Hk_JKGBAYT (Accessed on January 22, 2019)

Deep Learning Basics: Introduction and Overview - MIT

An introductory lecture overviewing the basics of deep learning including a few key ideas, subfields, and the big picture of why neural networks have inspired and energized an entire new generation of researchers. For more lecture videos visit our website or follow code tutorials on our GitHub repo. INFO: Website: https://deeplearning.mit.edu GitHub: https://github.com/lexfridman/mit-dee... Slides: http://bit.ly/deep-learning-basics-sl... Playlist: http://bit.ly/deep-learning-playlist OUTLINE: 0:00 - Introduction 0:53 - Deep learning in one slide 4:55 - History of ideas and tools 9:43 - Simple example in TensorFlow 11:36 - TensorFlow in one slide 13:32 - Deep learning is representation learning 16:02 - Why deep learning (and why not) 22:00 - Challenges for supervised learning 38:27 - Key low-level concepts 46:15 - Higher-level methods 1:06:00 - Toward artificial general intelligence

Source: https://www.youtube.com/watch?v=O5xeyoRL95U (Accessed on January 22, 2019)

Deep Learning State of the Art (2019) - MIT

New lecture on recent developments in deep learning is defined in the state of the art in our field (algorithms, applications, and tools). This is a complete list, but hopefully a good sampling of new exciting ideas. For more lecture videos visit our website or follow code tutorials on our GitHub repo. INFO: Website: https://deeplearning.mit.edu GitHub: https: //github.com/lexfridman/mit-dee ... Slides: http://bit.ly/2HiZyvP Playlist: http: // bit ly / deep-learning-playlist OUTLINE: 0:00 - Introduction 2:00 - BERT and Natural Language Processing 14:00 - Tesla Autopilot Hardware v2 +: Neural Networks at Scale 16:25- AdaNet: AutoML with Ensembles 18:32 - AutoAugment: Deep RL Data Augmentation 22:53 - Training Deep Networks with Synthetic Data 24:37 - Segmentation Annotation with Polygon-RNN ++ 26:39 - DAWNBench: Training Fast and Cheap 29:06 - BigGAN: State of the Art in Image Synthesis 30:14 - Video-to-Video Synthesis 32:12 - Semantic Segmentation 36:03 - AlphaZero & OpenI Five 43:34 - Deep Learning Frameworks 44:40 - 2019 and beyond CONNECT: - If you enjoyed this video, please subscribe to this channel. - Twitter: https://twitter.com/lexfridman-LinkedIn :https://www.linkedin.com/in/lexfridman

Source: https://www.youtube.com/watch?v=53YvP6gdD7U (Accessed on January 22, 2019)

Accelerating The Power of Deep Learning with Neural Networks and GPUs

Deep learning capabilities are driving powerful new opportunities for AI.

Deep learning using neural networks and Graphics Processing Units (GPUs) is starting to surpass machine learning for image recognition and other applications. Deep neural networks are helping to advance self-driving cars, faster development of new drugs, and real-time multiple-language translation for online chats.

Find out more about:

Machine learning versus deep learning
The power of inference
GPUs delivering near real-time performance

To read the full-text, visit the following URL:

https://www.tcs.com/content/dam/tcs/pdf/technologies/artificial-intelligence/abstract/Deep-learning-with-neural-networks-and-GPUs-0917-1.pdf

Source: https://www.tcs.com/accelerating-the-power-of-deep-learning (Accessed on January 22, 2019)

Monday, January 21, 2019

Riding the artificial intelligence wave

Artificial intelligence (AI) has been the talk of the start-up town for a while now. E-commerce major Flipkart plans to create an AI unit called AIForIndia. Online transportation network company Ola recently rolled out an AI-based real-time tracking mechanism for passenger safety. But Unicorns and established players aren't the only ones bullish on this technology. There is a clutch of newbie ventures that have focused their business models primarily around AI.

From leveraging artificial intelligence for those in search of a life partner to deploying it for curating gourmet and healthy food menus, start-ups are utilising the much-in-demand technology in the most unique ways.

AI offers tremendous growth potential for start-ups, with private sector investment scaling up over the years. As per industry estimates, more than $150 million has been invested across 400 AI-focused start-ups in India over the last few years. Meanwhile, private sector investments in AI in the US have been $18 billion. In Europe, they have totalled $2.6 billion and approximately $1.8 billion in Israel in the recent past.

Despite AI-focused investments in India appearing minuscule in front of the western countries, there is scope for growth for start-ups, say experts.

Sanmati Pande, founder and CEO, Growfitter, says, “AI helps in scaling up a business quickly irrespective of geographical boundaries. AI tools help to organise the raw data, which is compiled in specific formats and with deep learning techniques, provides detailed insights and statistics to stakeholders. AI-based start-ups are witnessing a keen interest from investors. AI can produce precise results with minimal resources and the scalability of the business can be achieved through an asset-light model with the help of AI techniques, which play a vital role for investors.”

Fitness and health-focused start-up Growfitter has built an AI-based preventive healthcare platform which is used by fitness centres, insurance companies and corporates. Their AI-powered app helps in tracking a customer's workout time and assess the fitness level, which helps in delving into the customer's health behaviour and generating a health score based on a proprietary algorithm. “Earlier, gyms witnessed renewal rates of around 25% and an average customer attendance of 100 days in a year. We have boosted the renewal rate to three times the current trend, while attendance has gone up to 175 days,'' says Pande.

A start-up like Betterhalf.ai is built around leveraging AI to help professionals find their life partners through compatibility scores based on multiple relationship dimensions and their interactions on the product. This marks a shift from traditional matchmaking which has been designed around physical, educational and caste-related traits of an individual. “We capture through AI an individual's personality in different relationship personality dimensions like emotional, social, intellectual, physical and values. We further curate through AI up to 10 relevant matches for them, cutting down 80% of the time they would have otherwise spent through traditional matchmaking,'' says Pawan Gupta, co-founder and CEO, Betterhalf.ai.

Meanwhile InnerChef, which curates ready-to-eat meals, healthy eats, gourmet experiences, indulgent desserts and detox cleanses, uses AI in building customised menus, for sales forecasting, capturing food trends and for consumer food recommendations.

Rajesh Sawhney, founder and CEO, InnerChef, says the venture has grown by 100% in the last one year. “We have plans to accelerate growth further and achieve three times revenue growth. We are currently in the National Capital Region, Bengaluru, Hyderabad and Mumbai. We are looking to expand our shared kitchen platform in new geographies this year.”

Growfitter, says Pande, witnessed attained $4.5 million in gross revenues in FY18, a 300% jump. “In the current fiscal, through the launch of our AI-based Incentivised Wellness Programme, our annualised revenue run rate for FY19 shall be $10 million. We have already launched operations in Indonesia and shall enter other South-East Asian countries soon.”

Betterhalf.ai, which has over 11,000 users across India, North America, Europe, South Africa and the UAE, is aiming to attain a user base of one million in the next two years. “10% users invite other users to the product. We expect this to shoot up to over 30% as the product and the AI matching algorithm becomes even more effective,'' says Gupta.

Source: https://www.dnaindia.com/business/report-riding-the-artificial-intelligence-wave-2710165 (Accessed on January 21, 2019)

Tiny, smart robots can swim through blood vessels to deliver drugs

tiny-smart-robots-can-swim-through-blood-vessels-to-deliver-drugs-1435196-2019-01-20 (Accessed on January 21, 2019)

Ensure all students have Aadhaar cards: Edu dept to Maha schools

SCHOOLS HAVE BEEN GIVEN THE TASK OF HANDING OUT AADHAAR CARD REGISTRATION FORMS TO PARENTS STARTING NEXT WEEK

MUMBAI: Schools across Maharashtra have been asked to ensure that no student is left without an Aadhaar card.

The education department recently received 816 machines from the government, which will be installed at schools across 408 talukas in the coming days.

The circular issued by the education department reads: “While the card is not required for school admissions, it is still an essential document for midday meal, scholarships, etc. It is thus important for students to get the card to avail these facilities.”

An official from the state education department said school staff will be trained for the task in the coming days.

Students who have an Aadhaar card but are yet to update their biometric details on the portal also need to complete the process soon.

Schools have been given the task of handing out forms to parents from the next week. After students register , schools have been asked to update their details on the government’s SARAL portal.

Prashant Redij, spokesperson of the Mumbai Principals Association, said the government should not shift the responsibility of ensuring that everyone has an Aadhaar card on schools.

“While 90%of students in most schools have the card, the others have not got them owing to some technical issues. It should be the responsibility of the government and the parents to ensure that children have their Aadhaar cards,” he added.

A suburban school principal said the initiative would add to the data entry work the school teachers have to do every day.

“When will we get the time to teach?” the principal added.

Source: Hindustan Times dated January 21, 2019

LIRC@SFIT