The majority of data scientists and machine learning practitioners use relational data in their work [State of ML and Data Science 2017, Kaggle, Inc.]. But training machine learning models on data stored in relational databases requires significant data extraction and feature engineering efforts. These efforts are not only costly, but they also destroy potentially important relational structure in the data. We introduce a method that uses Graph Neural Networks to overcome these challenges. Our proposed method outperforms state-of-the-art automatic feature engineering methods on two out of three datasets.

A core operation in reinforcement learning (RL) is finding an action that is optimal with respect to a learned state-action value function. This operation is often challenging when the learned value function takes continuous actions as input. We introduce deep RBF value functions: state-action value functions learned using a deep neural network with a radial-basis function (RBF) output layer. We show that the optimal action with respect to a deep RBF value function can be easily approximated up to any desired accuracy. Moreover, deep RBF value functions can represent any true value function up to any desired accuracy owing to their support for universal function approximation. By learning a deep RBF value function, we extend the standard DQN algorithm to continuous control, and demonstrate that the resultant agent, RBF-DQN, outperforms standard baselines on a set of continuous-action RL problems.

What does a neural network encode about a concept as we traverse through the layers? Interpretability in machine learning is undoubtedly important, but the calculations of neural networks are very challenging to understand. Attempts to see inside their hidden layers can either be misleading, unusable, or rely on the latent space to possess properties that it may not have. In this work, rather than attempting to analyze a neural network posthoc, we introduce a mechanism, called concept whitening (CW), to alter a given layer of the network to allow us to better understand the computation leading up to that layer. When a concept whitening module is added to a CNN, the axes of the latent space can be aligned with concepts of interest. By experiment, we show that CW can provide us a much clearer understanding for how the network gradually learns concepts over layers without hurting predictive performance.

The application of artificial intelligence in medicine and healthcare can assist with the optimization of the care trajectory of chronic disease patients; and it can suggest precision therapies for complex illnesses. Furthermore, algorithms can improve subject enrollment into clinical trials, leading to the creation of more efficacious medicines. To showcase the technology, three examples of artificial intelligence as applied to the medical field are examined. AI for lung cancer detection The science company Draper has begun a program designed to make artificial intelligence systems better at detecting lung cancer warning signs. This is by assessing medical images.

Ecosystem monitoring is central to effective management, where rapid reporting is essential to provide timely advice. While digital imagery has greatly improved the speed of underwater data collection for monitoring benthic communities, image analysis remains a bottleneck in reporting observations. In recent years, a rapid evolution of artificial intelligence in image recognition has been evident in its broad applications in modern society, offering new opportunities for increasing the capabilities of coral reef monitoring. Here, we evaluated the performance of Deep Learning Convolutional Neural Networks for automated image analysis, using a global coral reef monitoring dataset. The study demonstrates the advantages of automated image analysis for coral reef monitoring in terms of error and repeatability of benthic abundance estimations, as well as cost and benefit.

The intersection of artificial intelligence (AI) and GIS is creating massive opportunities that weren't possible before. AI, machine learning and deep learning are helping us make a better world by helping increase crop yield through precision agriculture, in fighting crime by deploying predictive policing models, or predicting when the next big storm will hit and being better equipped to handle it. One area of AI that has emerged in recent years is deep learning, and it has done exceedingly well in computer vision tasks. This is particularly useful for GIS, as satellite, aerial and drone imagery is being produced at a rate that makes it impossible to analyse and derive insight in a timely manner through traditional means. As the managing director of Esri's AI R&D Center in New Delhi, Rohit Singh leads the development of data science, deep learning and geospatial AI solutions in the ArcGIS platform.

A small Python 3 script to download public Tweets from a given Twitter account into a format suitable for AI text generation tools (such as gpt-2-simple for finetuning GPT-2). You can view examples of AI-generated tweets from datasets retrieved with this tool in the /examples folder. Inspired by popular demand due to the success of @dril_gpt2. The script is interacted via a command line interface. You can use this Colaboratory notebook (optimized from the original notebook for this use case) to train the model on your downloaded tweets, and generate massive amounts of Tweets from it.

We are thrilled to be collaborating with AWS across both edge and cloud-based solutions, covering the entire spectrum of AI. AWS DeepLens represents the latest addition to Intel's artificial intelligence solutions with Amazon which include the Intel Speech Enabled Developer Toolkit, the Intel powered Amazon Echo Show and Echo Look, and the Intel Xeon Scalable processors for AWS's new C5 instance family. C5 instances offer the lowest price per vCPU in the Amazon EC2 family and are ideal for running advanced compute-intensive workloads including machine and deep learning. The Intel AI portfolio has a broad suite of products and software solutions, from the edge to the cloud, enabling developers to focus on their science and drive the next wave of innovation in AI.

Sharing AI related news has become crucial in this era of digital transformation. Given the current wave, many AI researchers have turned into AI influencers to drive value and success to their respective field. These are the people who are driving conversations about AI across social media and other platforms. Please note: This is not a ranking article. Gregory Piatetsky is a well-known expert in Big Data, Business Analytics, Data Mining, Data Science, and Machine Learning and is among top influencers in those fields.

The world's tech powers are sending giant sums of money spinning into Canada, but while many see this as a sign of success, others are worried about researchers and intellectual property being swallowed wholesale. The country is in the midst of an artificial intelligence (AI) boom, with Google, Microsoft, Facebook, Huawei and other global heavyweights spending millions or even hundreds of millions of dollars on research hubs in Quebec, Ontario and Alberta. Canadian doors are open – some fear too open. Jim Hinton, an IP lawyer and founder of the Own Innovation consultancy, reckons that more than half of all AI patents in Canada end up being owned by foreign companies. What we need to be doing is getting money out of our ideas ourselves, instead of seeing foreign talent scoop it all up," said Hinton. "Otherwise we'll never have a Canadian champion." The country is home to hundreds of fledgling AI companies, including much-talked-about start-ups like Element AI and Deep Genomics, but they remain relatively small. "They don't have a strong market position yet," Hinton says. Deep learning pioneers such as Yoshua Bengio and Geoffrey Hinton (no relation to Jim) have nurtured top-notch talent in AI in Canada for years, back when AI was an emerging field. But despite Canadian inheriting this brilliant AI lead from the country's AI "godfathers", big foreign players have an unassailable advantage over homegrown efforts, Hinton said. "It's not an easy go for the average company to make a business out of AI.