SpecAugment: New and Simple Data Augmentation Technique for Audio Data

24 April 2019

SpecAugment: New and Simple Data Augmentation Technique for Audio Data

Researchers from Google AI have proposed a new data augmentation method for automatic speech recognition. The new method is based on a very simple idea and takes a new approach…

Algorithm Approximates Images Using Geometric Shapes

23 April 2019

Algorithm Approximates Images Using Geometric Shapes

Researchers from the Faculty of Technology, Art, and Design at Oslo Metropolitan University have presented an evolutionary art project where paintings are represented with overlapping geometric shapes. The proposed algorithm…

Facebook’s AI System Can Convert One Singer’s Voice Into Another

20 April 2019

Facebook’s AI System Can Convert One Singer’s Voice Into Another

Researchers from FAIR (Facebook AI Research) and Tel Aviv University presented an AI system that can convert one singer’s voice into another. In a pre-print paper published on arxiv, Eliya…

Springer Nature Published Its First Machine-generated Book

16 April 2019

Springer Nature Published Its First Machine-generated Book

Springer Nature published its first machine-generated book in chemistry. Under the name “Beta Writer” listed as the author of the book, lies an algorithm developed by scientists from Germany’s Goethe…

Gated-GAN: Style Transfer With Multiple Styles

15 April 2019

Gated-GAN: Style Transfer With Multiple Styles

A new method called Gated-GAN was proposed by Xinyuan Chen et al. and makes it possible to train multi-style Generative Adversarial Networks (GANs) for Style Transfer. A large number of variants…

Generating SVG Fonts with Deep Learning by Google Brain

15 April 2019

Generating SVG Fonts with Deep Learning by Google Brain

Researchers from Google Brain have proposed a new representation learning method for Scalable Vector Graphics (SVGs). They showed how the learned representations can be used to extract high-level characteristics of…

New Achor-Free Object Detection Method

11 April 2019

New Achor-Free Object Detection Method

Researchers from the National University of Defense Technology, Changsha in China, the Chinese Academy of Sciences and Inception Institute of Artificial Intelligence have proposed a novel method for object detection…

Neural Network Can Predict Code from Previous Edits

9 April 2019

Neural Network Can Predict Code from Previous Edits

Researchers from Google Brain have proposed a method for predicting code from previous edits. The novel method, based on deep neural networks is able to learn a sequence of code…

MIT Researchers Used Machine Learning To Improve Flavor of Basil

9 April 2019

MIT Researchers Used Machine Learning To Improve Flavor of Basil

Researchers from the Massachusetts Institute of Technology (MIT) have used machine learning to improve the flavor of lab-grown basil. In a new study, published in the journal PLOS One, researchers…

Neural Network Captures Special Video Moments in Google Photos

8 April 2019

Neural Network Captures Special Video Moments in Google Photos

A new method proposed by researchers at Google AI is able to identify “magic” moments in videos and create animations from them. As a new feature in Google Photos, the…

IBM Researchers Proposed New Biologically Plausible Learning Algorithm for Neural Networks

8 April 2019

IBM Researchers Proposed New Biologically Plausible Learning Algorithm for Neural Networks

Researchers from IBM Research have proposed a novel biologically plausible learning algorithm for neural networks. In their paper “Unsupervised Learning by Competing Hidden Units” published in the PNAS (Proceedings of…

Facebook AI Research Released PyTorch BigGraph

7 April 2019

Facebook AI Research Released PyTorch BigGraph

Facebook AI Research has announced a new open-source tool – PyTorch BigGraph. The new tool working on top of PyTorch enables training of multi-relation graph embeddings for very large graphs.…

Novel Video Frame Interpolation Method Based on Depth Maps

5 April 2019

Novel Video Frame Interpolation Method Based on Depth Maps

Researchers from Shanghai Jiao Tong Univers, University of California and Google, have proposed a novel method for video frame interpolation based on depth information. Video frame interpolation has been a…

FastFCN: Novel State-of-the-art Method for Semantic Segmentation

5 April 2019

FastFCN: Novel State-of-the-art Method for Semantic Segmentation

Researchers from the Chinese Academy of Sciences and Deepwise AI Lab, have proposed a new method for semantic segmentation which reduces computational complexity and achieves state-of-the-art results. Semantic segmentation, as…

“GAN You Do the GAN GAN?”: Modeling GANs with a GAN

4 April 2019

“GAN You Do the GAN GAN?”: Modeling GANs with a GAN

Joseph Suarez, a Stanford grad student has released a project where he tried to train a GAN to model a distribution of GANs. In his short paper named “GAN You…

TossingBot – Unifying Physics and Deep Learning To Learn Tossing

30 March 2019

TossingBot – Unifying Physics and Deep Learning To Learn Tossing

Researchers from Google, Princeton, Columbia University and MIT unified physics and deep learning to develop a tossing robot called, TossingBot. Grasping has been one of the most famous challenges in…

Two New Exoplanets Discovered Using Deep Learning and Kepler’s Data

30 March 2019

Two New Exoplanets Discovered Using Deep Learning and Kepler’s Data

For the first time in history, an artificial intelligence system has discovered exoplanets using Kepler’s K2 observational data. Researchers from Google have discovered two new exoplanets using deep learning and…

MIT’s Human-Centered Autonomous Vehicle

27 March 2019

MIT’s Human-Centered Autonomous Vehicle

Lex Fridman, a researcher from the Massachusetts Institute of Technology (MIT) has proposed a new human-centered autonomous vehicle system. The idea behind the concept of human-centered autonomous vehicle is to…

OpenAI Released New Implicit Image Generation Methods for Energy-Based Models

25 March 2019

OpenAI Released New Implicit Image Generation Methods for Energy-Based Models

Researchers from OpenAI have proposed new implicit data generation methods for Energy-Based Models. Energy-Based Models (EBMs) are able to discover data dependencies by applying a measure of compatibility (called energy)…

New Method from Google AI: High-Fidelity Image Generation with Fewer Labels

25 March 2019
stylegan

New Method from Google AI: High-Fidelity Image Generation with Fewer Labels

A new method developed by researchers from Google AI is achieving state-of-the-art results in high-fidelity image generation with GANs using 10 times fewer labels. Generative Adversarial Networks have done huge…

Google’s Stadia Uses Style Transfer ML To Change a Game’s Art in Real-time

22 March 2019

Google’s Stadia Uses Style Transfer ML To Change a Game’s Art in Real-time

Google announced the launch of Stadia – a virtual stadium where games can be played in the cloud. In this year’s GDC (Game Developers Conference) this big announcement meant that…