In this longish post, I have tried to explain Deep Learning starting from familiar ideas like machine learning. This approach forms a part of my forthcoming book. You can connect with me on Linkedin to know more about the book. I have used this approach in my teaching. It is based on'learning by exception,' i.e. understanding one concept and it's limitations and then understanding how the subsequent concept overcomes that limitation. We thus develop a chain of thought that starts with linear regression and extends to multilayer perceptron (Deep Learning).

The Alexa Shopping: Privacy and Machine Learning team in Cambridge is looking for Applied Science Interns for summer 2020. As an applied scientist in Alexa Shopping, you will be responsible for the research, design and development of new machine learning and privacy enhancing technologies for voice-based and multimodal shopping experiences. You will be working with top scientists and engineers, as well as with product teams and other research partners, both locally and abroad. Your work will combine data mining, systems and software development, exploration of new technologies, as well as working towards publications and presentations at top scientific conferences. The team has expertise in Gaussian processes, approximate inference, probabilistic modelling, deep learning, crowdsourcing, user modelling, differential privacy, and other related fields.

Today, we're joined by Aydogan Ozcan, Professor of Electrical and Computer Engineering at UCLA, where his research group focuses on photonics and its applications to nano- and biotechnology. In our conversation, we explore his group's research into the intersection of deep learning and optics, holography and computational imaging. We specifically look at a really interesting project to create all-optical neural networks which work based on diffraction, where the printed pixels of the network are analogous to neurons. We also explore some of the practical applications for their research and other areas of interest for their group. "More On That Later" by Lee Rosevere licensed under CC By 4.0

When it comes to AI (Artificial Intelligence), VCs (venture capitalists) continue to be aggressive with their fundings. During the third quarter, 965 AI-related companies in the US raised a total of $13.5 billion. In fact, this year should see a record in total fundings (last year's total came to $16.8 billion). Some of the deals have been, well, staggering. Just look at the $1 billion that Microsoft shelled out for an equity stake in OpenAI (the company is one of the few that is pursuing Strong AI). So what has been the result of all this activity? What have been the breakthroughs for AI this year?

TensorFlow 2.0 was made available in October. One of the major highlights of this release was the integration of Keras into TensorFlow. Keras is an open-source deep-learning library that is designed to enable fast, user-friendly experimentation with deep neural networks. It serves as an interface to several deep learning libraries, most popular of which is TensorFlow, and it was integrated into TensorFlow main codebase in TensorFlow 2.0. In September, Lex Fridman, Research scientist at MIT popularly known for his podcasts, spoke to François Chollet, who is the author of Keras on Keras, Deep Learning, and the Progress of AI.

Check out the entire Machine Learning Playlist: https://bit.ly/2NG9tK4

Counting and identifying characteristics of crowds can provide organizations with a lot of valuable insights. Yet challenges like image distortion, density, and different camera angles can make analyzing images accurately very challenging. Ganes Kesari, Co-founder and Head of Analytics at Gramener, joins the Intel on AI podcast to discuss how Gramener has created a crowd counting solution that can overcome those challenges and produce a very rapid and accurate analysis of images. He talks about how Gramener has utilized this solution for several AI for good projects including a joint effort with Microsoft* to count Antarctic penguin colonies. Ganes explains how their solution used convolutional neural networks (CNNs) using density-based estimations to deliver a more accurate penguin count than traditional manual counting methods.

A child who has never seen a pink elephant can still describe one -- unlike a computer. "The computer learns from data," says Jiajun Wu, a PhD student at MIT. "The ability to generalize and recognize something you've never seen before -- a pink elephant -- is very hard for machines." Deep learning systems interpret the world by picking out statistical patterns in data. This form of machine learning is now everywhere, automatically tagging friends on Facebook, narrating Alexa's latest weather forecast, and delivering fun facts via Google search. But statistical learning has its limits.

There's a huge disadvantage of sliding windows detection which is the Computational cost, because we're cropping out so many different square regions in the image and running each of them independently through a \(convnet \). If we use the very course stride, a very big stride, very big step size, then that would reduce the number of windows we need to pass through the \(convnet \), but that coarser granularity may hurt performance, whereas if we use a very fine granularity or a very small stride then the huge number of all these little regions we're passing through the \(convnet \) means that there's a very high computational cost. Before the rise of neural networks people used to use much simpler classifiers, like a simple linear classifier overhand engineer features in order to perform object detection, and in that error because each classifier was relatively cheap to compute it was just a linear function, sliding windows detection ran properly, it was not a bad method, but with \(convnets \) now running a single classification task is much more expensive and sliding windows this way is infeasible slow. Unless we use a very fine granularity or a very small stride we end up not able to localize the objects that accurately within the image as well.

TensorFlow 2.0, released in October, has got developers excited about a myriad of features and its ease of use. At the EuroPython Conference 2019, Brad Miro, developer programs engineer at Google talked about the updates being made to TensorFlow 2.0. He also gave an overview of how Google is using TensorFlow, moving on to why Python is important for TensorFlow development and how to migrate from TF 1.x to TF 2.0. EuroPython is one of the most popular Python programming language community conferences. Below are some highlights from Brad's talk at EuroPython.