Patient notes contain a wealth of information of potentially great interest to medical investigators. However, to protect patients' privacy, Protected Health Information (PHI) must be removed from the patient notes before they can be legally released, a process known as patient note de-identification. The main objective for a de-identification system is to have the highest possible recall. Recently, the first neural-network-based de-identification system has been proposed, yielding state-of-the-art results. Unlike other systems, it does not rely on human-engineered features, which allows it to be quickly deployed, but does not leverage knowledge from human experts or from electronic health records (EHRs). In this work, we explore a method to incorporate human-engineered features as well as features derived from EHRs to a neural-network-based de-identification system. Our results show that the addition of features, especially the EHR-derived features, further improves the state-of-the-art in patient note de-identification, including for some of the most sensitive PHI types such as patient names. Since in a real-life setting patient notes typically come with EHRs, we recommend developers of de-identification systems to leverage the information EHRs contain.

It's not about what it can do, but the effects of its prioritization As buzzwords become ubiquitous they become easier to tune out. We've finely honed this defense mechanism, for good purpose. It's better to focus on what's in front of us than the flavor of the week. CRISPR might change our lives, but knowing how it works doesn't help you. VR could eat all media, but it's hardware requirements keep it many years away from common use.

Researchers from the Google Brain deep learning project have already taught AI systems to make trippy works of art, but now they're moving on to something potentially darker: AI-generated, human-independent encryption. According to a new research paper, Googlers Martín Abadi and David G. Andersen have willingly allowed three test subjects -- neural networks named Alice, Bob and Eve -- to pass each other notes using an encryption method they created themselves. As the New Scientist reports, Abadi and Andersen assigned each AI a task: Alice had to send a secret message that only Bob could read, while Eve would try to figure out how to eavesdrop and decode the message herself. The experiment started with a plain-text message that Alice converted into unreadable gibberish, which Bob could decode using cipher key. At first, Alice and Bob were apparently bad at hiding their secrets, but over the course of 15,000 attempts Alice worked out her own encryption strategy and Bob simultaneously figured out how to decrypt it.

Microsoft this week released an updated version of its Microsoft Cognitive Toolkit as an open source Beta. The deep learning system is used to speed advances in areas such as speech and image recognition and search relevance on CPUs and Nvidia GPUs. It also works with Microsoft's Azure GPU offering. The Microsoft computer scientists who developed the toolkit initially were looking for a tool to speed up and improve their own research. Initially called "Microsoft/CNTK," it morphed into an offering that Microsoft customers and flagship product groups depend on for a wide variety of deep learning tasks, the company said.

This year, the Association for Computing Machinery (ACM) celebrates 50 years of the ACM Turing Award, the most prestigious technical award in the computing industry. The Turing Award, generally regarded as the'Nobel Prize of computing', is an annual prize awarded to "an individual selected for contributions of a technical nature made to the computing community". In celebration of the 50 year milestone, renowned computer scientist Melanie Mitchell spoke to CBR's Ellie Burns about artificial intelligence (AI) – the biggest breakthroughs, hurdles and myths surrounding the technology. MM: There are many important examples of AI in the mainstream; some very visible, others blended in so well with other methods that the AI part is nearly invisible. Web search is an "invisible" example that has had perhaps the broadest impact.

A team at Google has built a system to show that artificial intelligence can build its own form of encryption. While not very complex currently, this research could set the table for encryption that gets stronger as hackers attempt to crack it. To see if the artificial intelligence could learn to encrypt on its own, the AI researchers at Google Brain, a unit of the search company focused on deep learning, built a game with three entities powered by deep neural networks: Alice, Bob, and Eve. Alice was designed to send an encrypted message of 16 zeroes and ones to Bob, which was designed to decrypt the message. The two bots started with a shared key, a foundation for the message's encryption.

We've paid a lot of attention lately to Convolutional Neural Nets (CNNs) as the cornerstone of 2nd gen NNs and spent some time on Spiking Neural Nets (SNNs) as the most likely path forward to 3rd gen, but we'd really be remiss if we didn't stop to recognize Recurrent Neural Nets (RNNs). Because RNNs are solid performers in the 2nd gen NN world and perform many tasks much better than CNNs. These include speech-to-text, language translation, and even automated captioning for images. By count, there are probably more applications for RNNs than for CNNs. On one scale RNNs have much more in common with the larger family of NNs than do CNNs which have very unique architecture. RNNs allow inputs of strings of data to be assessed together and those strings can be of widely varying lengths.

Tensorflow is a deep learning software developed by Google as a successor to its DistBelief software, which also focused on deep learning. Deep learning is a concept where activity of neurons in the brain is emulated. The basic idea is to mimic the activity of neurons in the brain in the context of machines and software, through the development of an artificial neural network. This network could help machines and software to perform an array of tasks ranging from speech recognition, image recognition etc. Tensorflow is software built around these principles. The software was built to reason and think like humans and this was made possible by feeding it a vast amount of data and then training it to recognize objects and patterns from the data.

While the number of headlines about machine learning might lead one to think that we just discovered something profoundly new, the reality is that the technology is nearly as old as computing. It's no coincidence that Alan Turing, one of the most influential computer scientists of all time, started his 1950 treatise on computing with the question "Can machines think?" From our science fiction to our research labs, we have long questioned whether the creation of artificial versions of ourselves will somehow help us uncover the origin of our own consciousness, and more broadly, our role on earth. Unfortunately, the learning curve on AI is really damn steep. By tracing a bit of history, we should hopefully be able to get to the bottom of wtf machine learning really is.

Actually I've known about MXnet for weeks as one of the most popular library / packages in Kaggler, but just recently I heard bug fix has been almost done and some friends say the latest version looks stable, so at last I installed it. I think that the most important feature of MXnet is its implementation of not only Deep Neural Network (DNN) but also Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) in R, because as far as I've known there has been no R packages implementing CNN (and/or RNN). In the original post of my blog, I tried a CNN {mxnet} R package with a short version of MNIST handwritten digit datasets whose maximum accuracy may be less than 0.98 for its small sample size. As a result, CNN of {mxnet} performed accuracy 0.976: this is better than Random Forest (0.951), Xgboost (0.953) or DNN by {h2o} (0.962).