You can hardly turn on the television news, pull a magazine off a rack in a doctor's office, or check out your social media without being confronted by a discussion about artificial intelligence. Whether the writer or talking head is decrying the imminent robot apocalypse or celebrating our deep-learning-based salvation, most of the coverage has one thing in common: an imprecise definition of AI. AI is, at its base, nothing more than software that simulates intelligence. One specific type of AI is cropping up all around the Internet: conversational AI, mostly in the form of chatbots. The most recent and high-profile news about AI was Google's announcement that its AI, called Google Assistant, beat the Turing test--150 times.

Description: Learn how to Visualize Data with MATLAB in 2D, 3D, 4D, 5D and create animated plots with tens of projects and examples!

Cyclists and pedestrians move along Trinity Street past St Johns College, part of the University of Cambridge in Cambridge, U.K., on Sunday, Oct. 5, 2014. As the U.K. government cuts funding for universities, Cambridge's colleges are becoming increasingly entrepreneurial about making money. DeepMind, a Google-owned lab that is at the front of the AI race, is gifting the University of Cambridge an undisclosed sum to appoint a "DeepMind Chair of Machine Learning". The donation comes after DeepMind and other tech companies have been criticised for poaching dozens of leading academics from top universities like Cambridge, luring them with big salaries that universities can't compete with. Here's a report about some of DeepMind's previous Oxbridge hires.

The great strength of CNN architectures is their capability to automatically learn a hierarchy of feature detectors in order to solve some task. What it has been observed is that regardless of the architecture, the dataset and the target semantic space (and of course the initialization assuming it's random), the first layers seem to always converge to specific kinds of feature detectors: the Gabor Filters. This is actually a very interesting and important phenomenon as it seems to suggest the Gabor Filters are the most efficient way to start the semantic extraction process from an image. It would mean Gabor Filters block is a sort of "generic building block" which could be used to design NN aimed at solving computer vision problems This is one of the main goal of Transfer Learning: finding "building blocks" which can be composed to build a NN and fine tuned, instead of trained from scratch, on the Dataset Being able to properly understand how the CNN specializes while training is important to get to Transfer Learning: "transferring" the network "capability of solving a problem" which means basically adapting its weights properly, to another similar problem in a Data Efficient Way The data efficiency is in fact one of the most important aspects of transfer learning: it is well known that Supervised Learning is an effective way to make a certain, typically big, NN become able to solve a problem but it scales badly in terms of data as it typically requires A LOT OF supervision signal which, in case of manual annotation, is expensive to collect as it relies on humans to provide it. Furthermore the more the task is difficult, the more the annotations need to be provided by human experts instead of normal people and the former one's time is more expensive than the latter ones

Deep learning is an important part of the business of Google, Amazon, Microsoft, and Facebook, as well as countless smaller companies. It has been responsible for many of the recent advances in areas such as automatic language translation, image classification, and conversational interfaces. We haven't gotten to the point where there is a single dominant deep learning framework. TensorFlow (Google) is very good, but has been hard to learn and use. Also TensorFlow's dataflow graphs have been difficult to debug, which is why the TensorFlow project has been working on eager execution and the TensorFlow debugger.

Where do I return these shoes? Is Jane or Bob a better hire for our sales department? Just a few years ago, it took a human to answer these queries. Today, machines do it faster and -- more often than not -- better. The growth of human-facing, AI-based tools has given rise to a new set of niche, midmarket companies harnessing the advanced technology for B2B and B2C applications.

Contextual multi-armed bandit problems arise frequently in important industrial applications. Existing solutions model the context either linearly, which enables uncertainty driven (principled) exploration, or non-linearly, by using epsilon-greedy exploration policies. Here we present a deep learning framework for contextual multi-armed bandits that is both non-linear and enables principled exploration at the same time. We tackle the exploration vs. exploitation trade-off through Thompson sampling by exploiting the connection between inference time dropout and sampling from the posterior over the weights of a Bayesian neural network. In order to adjust the level of exploration automatically as more data is made available to the model, the dropout rate is learned rather than considered a hyperparameter. We demonstrate that our approach substantially reduces regret on two tasks (the UCI Mushroom task and the Casino Parity task) when compared to 1) non-contextual bandits, 2) epsilon-greedy deep contextual bandits, and 3) fixed dropout rate deep contextual bandits. Our approach is currently being applied to marketing optimization problems at HubSpot.

For different language pairs, word-level neural machine translation (NMT) models with a fixed-size vocabulary suffer from the same problem of representing out-of-vocabulary (OOV) words. The common practice usually replaces all these rare or unknown words with a 〈UNK〉 token, which limits the translation performance to some extent. Most of recent work handled such a problem by splitting words into characters or other specially extracted subword units to enable open-vocabulary translation. Byte pair encoding (BPE) is one of the successful attempts that has been shown extremely competitive by providing effective subword segmentation for NMT systems. In this paper, we extend the BPE style segmentation to a general unsupervised framework with three statistical measures: frequency (FRQ), accessor variety (AV) and description length gain (DLG). We test our approach on two translation tasks: German to English and Chinese to English. The experimental results show that AV and DLG enhanced systems outperform the FRQ baseline in the frequency weighted schemes at different significant levels.

The challenge of assigning importance to individual neurons in a network is of interest when interpreting deep learning models. In recent work, Dhamdhere et al. proposed Total Conductance, a "natural refinement of Integrated Gradients" for attributing importance to internal neurons. Unfortunately, the authors found that calculating conductance in tensorflow required the addition of several custom gradient operators and did not scale well. In this work, we show that the formula for Total Conductance is mathematically equivalent to Path Integrated Gradients computed on a hidden layer in the network. We provide a scalable implementation of Total Conductance using standard tensorflow gradient operators that we call Neuron Integrated Gradients. We compare Neuron Integrated Gradients to DeepLIFT, a pre-existing computationally efficient approach that is applicable to calculating internal neuron importance. We find that DeepLIFT produces strong empirical results and is faster to compute, but because it lacks the theoretical properties of Neuron Integrated Gradients, it may not always be preferred in practice. Colab notebook reproducing results: http://bit.ly/neuronintegratedgradients

We propose a framework for the completely unsupervised learning of latent object properties from their interactions: the perception-prediction network (PPN). Consisting of a perception module that extracts representations of latent object properties and a prediction module that uses those extracted properties to simulate system dynamics, the PPN can be trained in an end-to-end fashion purely from samples of object dynamics. The representations of latent object properties learned by PPNs not only are sufficient to accurately simulate the dynamics of systems comprised of previously unseen objects, but also can be translated directly into human-interpretable properties (e.g., mass, coefficient of restitution) in an entirely unsupervised manner. Crucially, PPNs also generalize to novel scenarios: their gradient-based training can be applied to many dynamical systems and their graph-based structure functions over systems comprised of different numbers of objects. Our results demonstrate the efficacy of graph-based neural architectures in object-centric inference and prediction tasks, and our model has the potential to discover relevant object properties in systems that are not yet well understood.