AI Technology & Industry Review
Technical review of the newest machine intelligence research.
The MXNet team and the Amazon Web Services AI lab recently teamed up with New York University / NYU Shanghai to announce Deep Graph Library (DGL), a Python package that provides easy implementations of GNNs research. Some fast takeaways on DGL:
Insilico Medicine, a drug discovery startup located in Johns Hopkins University, has introduced MOSES (Molecular Sets), a platform which can be used to compare model accuracy in molecular generation. MOSES provides a standardized benchmarking dataset, a set of open-sourced models with unified implementation, and evaluation metrics.
The digital painting tool GANpaint has gone viral on social media. The product of a team of high-profile researchers from MIT, IBM, Google, and the Chinese University of Hong Kong, GAPpaint allows anyone — even those with little knowledge of digital painting or photoshop — to “paint” incredibly complex and detailed photorealistic scenes.
In 2016 Google’s DeepMind stunned the world when their Go computer AlphaGo secured a historic victory over Korean grandmaster Lee Sedol. Yesterday the UK’s top AI team delivered their latest “wow moment” as their AI system AlphaFold topped the Critical Assessment of Structure Prediction (CASP) competition.
Jordan believes each of these should be resolved separately and on their own merits, and that a human-imitative AI system should not be the principal strategy for approaching IA and II problems.
Another video game has succumbed to the strength of artificial intelligence. Uber researchers announced yesterday that their AI has completely solved Atari’s Montezuma’s Revenge, a classic game that involves moving a character from one room to another while killing enemies and collecting jewels in a 16th century Aztec-like pyramid.
ImageNet Pre-training is common in a variety of CV (Computer Vision) tasks, as something of a consensus has emerged that pre-training can help a model learn transferrable information that can be useful for target tasks.
DARCCC (Detecting Adversaries by Reconstruction from Class Conditional Capsules) is a technique which uses a similarity metric to compare reconstructed images with an original input image to identify whether it was an adversarial image, and further detects whether the system was attacked.
Researchers from Japanese electronics giant Sony have trained the ResNet-50 neural network model on ImageNet in a record-breaking 224 seconds — 43.4 percent better than the previous fastest time for the benchmark task.
Deep Learning has become an essential toolbox which is used in a wide variety of applications, research labs, industries, etc. In this tutorial given at NIPS 2017, the speakers provide a set of guidelines which will help newcomers to the field understand the most recent and advanced models and their application to diverse data modalities.
New research from Carnegie Mellon University, Peking University and the Massachusetts Institute of Technology shows that global minima of deep neural networks can been achieved via gradient descent under certain conditions. The paper Gradient Descent Finds Global Minima of Deep Neural Networks was published November 12 on arXiv.
Analog quantum computing has been an appealing tool with potential real-life applications on various optimization and simulation tasks, and much less stringent requirements on error corrections compared to universal quantum computing.
Researchers from the University of California, Santa Barbara, and the University of Chicago have published a paper which identifies the risk of bad actors using smartphones’ WiFi signals to “see” through walls and surreptitiously track humans in their private rooms and offices.
If you’ve ever wondered whether Dota 2 or League of Legends is the most popular multiplayer online battle arena game, or how long you’d need to spend on a treadmill to burn off that party size bag of chips you just ate, you know that you can probably find the answer by accessing a couple of relevant information sources and then applying what seems like a natural and straightforward reasoning process.
Facebook announced today that it is open-sourcing QNNPACK, a high-performance kernel library optimized for mobile AI. The computing power of mobile devices is but a tiny fraction of that of data center servers. As such it is essential to find ways to optimize mobile devices’ hardware performance in order to run today’s compute-hungry AI applications.
Information retrieval (IR) is the activity of retrieving information from a collection of sources stored on computers, based on user queries. IR enjoys a history of one century [1], and serves as the heart of many ubiquitous applications such as web search, product recommendation, and personal feeds on social networks.
It’s known that deep neural network models — mostly discriminative models — are prone to making poor predictions with high confidence when they encounter data distribution scenarios that vary too much from their training data.
A new paper from Julia Computing Co-Founder and CTO Keno Fischer and Senior Research Engineer Elliot Saba introduces a method and implementation for offloading sections of Machine Learning models written in Julia programming language to TPUs.
DeepMind announced today that it has opened its Graph Nets (GN) library to the public, enabling the use of graph networks in TensorFlow and Sonnet. Graph Nets is a machine learning framework that was published by DeepMind, Google Brain, MIT and University of Edinburgh on Jun 15.
The DeepMimic paper’s first author, Berkeley PhD student Xue Bin Peng, has now open-sourced the project’s codes, data, and frameworks. Moreover, Peng’s new research demonstrates that DeepMimic’s simulated characters can also learn to perform highly dynamic movements by using regular video clips of human examples as input data.
DeepMind announced today that it is open-sourcing its TRFL (pronounced ‘truffle’) library, which contains a variety of building blocks useful for developing reinforcement learning (RL) agents in TensorFlow.
Founded in 1999, Tokyo-based DeNA has developed popular platforms and services for gaming, E-commerce, automotive, healthcare and entertainment content distribution. As AI continues transforming all things digital, DeNA is expanding its deep learning tech capabilities to support R&D on new techniques.
A Shanghai Jiao Tong University research team has announced the world’s first software for photonic analog quantum computing and simulation. “Feynman Photonic Analog Quantum Simulation” (FeynmanPAQS) is named after renown quantum physicist Richard P. Feynman.
UC Berkeley researchers have published a paper demonstrating how Deep Reinforcement Learning can be used to control dexterous robot hands for complicated tasks. Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations proposes a low-cost and high-efficiency control method that uses demonstration and simulation techniques to accelerate the learning process.
In conjunction with yesterday’s release of open source AI software framework PyTorch 1.0, leading deep learning course developer Fast.ai has announced its first open source library for deep learning — fastai v1.
“Best GAN samples ever yet? Very impressive ICLR submission! BigGAN improves Inception Scores by >100.” The above Tweet is from renowned Google DeepMind research scientist Oriol Vinyals. It was retweeted last week by Google Brain researcher and “Father of Generative Adversarial Networks” Ian Goodfellow, and picked up momentum and praise from AI researchers on social media.
Semiconductor giant Arm today unveiled the Cortex-A76AE, a microarchitecture CPU designed for use in autonomous vehicles. The UK-based Softbank subsidiary hailed the new chip as the world’s first autonomous-class processor with integrated safety features.
Georgia Tech and Google Brain researchers have introduced the new interactive tool GAN Lab, which visually presents the training process of complex machine learning model Generative Adversarial Networks (GANs). Even machine learning newbs can now experiment with GAN models using only a common web browser.
Petuum and Carnegie Mellon University (CMU) researchers have introduced Texar, an open-source general purpose text generation toolkit that can help boost R&D in fast model prototyping and experimentation.
Microsoft today announced a new neural text-to-speech synthesis system that makes computer voices nearly indistinguishable from human recordings. Neural TTS currently available for preview through Azure Cognitive Services Speech Services.
Deep Reinforcement Learning has shown impressive performance in a wide range of applications, including video games. StarCraft II, one of the most challenging Real Time Strategy (RTS) games, has however remained unsolved. Until now.
In a new paper Durham University researchers introduce a anomaly detection model, GANomaly, comprising a conditional generative adversarial network that “jointly learns the generation of high-dimensional image space and the inference of latent space.” The process enables the model to perform anomaly detection tasks even in sample-poor environments.
Chinese Internet mogul Jack Ma has a flair for naming new businesses: Alibaba originates from a character made famous in the One Thousand and One Nights collection of Arabian folk tales; while the company’s R&D arm Damo Academy derives from the name of a Chinese Buddhist monk instrumental in the creation of Shaolin Kung Fu.
Researchers from MIT, Google, and Xian Jiaotong University recently published a paper proposing AutoML for Model Compression (AMC), which leverages reinforcement learning to shorten model compression processing time and improve results.
Google AI lead Jeff Dean recently posted a link to his 1990 senior thesis on Twitter, which set off a wave of nostalgia for the early days of machine learning in the AI community. Parallel Implementation of Neural Network Training: Two Back-Propagation Approaches may be almost 30 years old and only eight pages long, but the paper does a remarkable job of explaining the methods behind neural network training and the modern development of artificial intelligence.
Researchers from Stanford University and University of California Berkeley have introduced Gibson Environment, a real-world-based virtual environment for training and testing active perception agents.
Compositing multiple objects from different sources into one image is a formidable computer graphics challenge involving relative scaling, spatial layout, occlusion, and viewpoint transformation of objects from the 3D world to a 2D image.
The “cocktail party effect” describes humans’ ability to hold a conversation in a noisy environment by listening to what their conversation partner is saying while filtering out other chatter, music, ambient noises, etc.
Leading Chinese AI startup 4Paradigm has announced a strategic partnership with Shanghai’s Ruijin Hospital on AI application in healthcare, especially chronic health conditions.
Google Brain Research Scientist Ian Goodfellow has tweeted an alarm about IoT hacking of a particularly nightmarish type, after Brown University security researchers were able to remotely access and control a robot in a university research lab. The research also showed that many robotic labs worldwide may be vulnerable to such a takeover technique.