AI Technology & Industry Review
A Facebook AI research team explores strategies for convolutional neural network scaling, aiming to provide a framework for analyzing scaling strategies under various computational constraints.
UC Berkeley and Adobe Research have introduced a “universal” detector that can distinguish real images from generated images regardless of what architectures and/or datasets were used for training.
Researchers proposed a “radioactive data” technique for subtly marking images in a dataset to help researchers later determine whether they were used to train a particular model.
Google researchers have introduced a new face detection framework called BlazeFace, adapted from the Single Shot Multibox Detector (SSD) framework and optimized for inference on mobile GPUs.
Synced Global AI Weekly May 5th
Designing accurate and efficient CNNs for mobile devices is challenging due to the large design space and expensive computational methods. Although many mobile CNNs are available for developers to train and deploy to mobile devices, existing CNN architecture may not be able to achieve the best results for some tasks on mobile devices.
Researchers from Facebook, the National University of Singapore, and the Qihoo 360 AI Institute have jointly proposed OctConv (Octave Convolution), a promising new alternative to traditional convolution operations. Akin to a “compressor” for Convolutional Neural Networks (CNN), the OctConv method saves computational resources while boosting effectiveness.
It is no secret that deep neural networks (DNNs) can achieve state-of-the-art performance in a wide range of complicated tasks. DNN models such as BigGAN, BERT, and GPT 2.0 have proved the high potential of deep learning. Deploying DNNs on mobile devices, consumer devices, drones and vehicles however remains a bottleneck for researchers.
Researchers from NLP startup Shannon.AI have published a study proposing Glyce — a set of Chinese glyph-vectors for Chinese character representations. Glyce has already mastered 13 core Chinese natural language processing tasks.
In his 1988 IEEE paper Cellular Neural Networks: Theory, UC Berkeley PhD student Lin Yang proposed Cellular Neural Network theory, a predecessor of the Convolutional Neural Networks (CNN) that would later revolutionize machine learning. Based on this theory, Yang blueprinted a 20*20 parallel simulated circuit chip in the university lab.
The internet loves those little looping action images we call GIFs. They can tell a short visual story in a small file size that has high portability. The visual quality of GIFs is however usually low compared to the videos they were sourced from. If you are sick of fuzzy, low resolution GIFs, then researchers from Stony Brook University, UCLA, and Megvii Research have just the thing for you: “the first learning-based method for enhancing the visual quality of GIFs in the wild.”
The proliferation of social media in our daily lives has profoundly changed the way we work and play with others. It has also created an entirely new job: thousands of people worldwide now work for Google, Facebook and Twitter “Community Operations Teams.” Whenever a user flags content as offensive, it’s sent to these guys for review.
Text-based CAPTCHA remain one of the most visible and commonly used mechanisms for website security. As a sort of online gatekeeper that distinguishes between humans and bots, the little solvable image fields have critical commercial applications in blocking automatic spam and preventing e-transfer fraud; and can also stop bots from spreading fraudulent information, etc.
The dearth of AI talents capable of manually designing neural architecture such as AlexNet and ResNet has spurred research in automatic architecture design. Google’s Cloud AutoML is an example of a system that enables developers with limited machine learning expertise to train high quality models. The trade-off, however, is AutoML’s high computational costs.
From Hayao Miyazaki’s Spirited Away to Satoshi Kon’s Paprika, Japanese anime has made it okay for adults everywhere to enjoy cartoons again. Now, a team of Tsinghua University and Cardiff University researchers have introduced CartoonGAN — an AI-powered technology that simulates the styles of Japanese anime maestri from snapshots of real world scenery.
New York-based AI startup LexSet.ai wants to make furniture shopping as simple and fun as playing a kid’s video game, with an AI-powered plug-in interior designer that “analyzes your home…
According to an AWS blog post, Apache MXNet now supports Keras 2 and the Keras-MXNet deep learning backend is now available to developers.
Cambricon today unveiled its Cambricon 1M chip for edge computing, and the MLU100, the first in a new chip series for cloud computing.
To boost learning research aimed at endowing robots with better generalization capabilities, Yi Wu from UC Berkeley and Yuxin Wu, Georgia Gkioxari, and Yuandong Tian from Facebook AI research recently published the paper Building Generalizable Agents with a Realistic and Rich 3D Environment.
To boost CNN performance, industry leaders and startups alike are now racing to develop specialized AI chips.
Human-machine interaction is rapidly evolving. Today, 12% of Chinese users opt for voice input over typing. What does this signal for tomorrow?
Dileep George and D. Scott Phoenix are confident they’re on the right path to achieve AGI with unique solutions combining machine learning and the human brain itself.
Today, the “Godfather of Deep Learning” Dr. Geoffrey Hinton published the paper Dynamic Routing Between Capsules on Arxiv. This is being hailed as a revolutionary work that could totally revamp computer vision research.
Facebook AI Researcher Yuandong Tian says new AlphaGo Zero paper is destined to be a classic.
In the article “Memory, attention, sequences”, the author predicts that future work on neural networks will emphasize understanding complex spatio-temporal data from the real world, which is highly contextual and noisy.
The ShuffleNet utilizes pointwise group convolution and channel shuffle to reduce computation cost while maintaining accuracy. It manages to obtain lower top-1 error than the MobileNet system on ImageNet classification, and achieves ~13x actual speedup over AlexNet while maintaining comparable accuracy.
This method, with considerable computation time and controllable model size, is particularly simple and very useful for implementations on mobile devices such as phones, drones, etc.
Compared to SMT, NMT can train multiple features jointly and does not need prior domain knowledge, enabling zero-shot translation. In addition to higher BLEU score and better sentence structure, NMT can also help reduce morphology errors, syntax errors, and word order errors of SMT.
We interviewed Mr. Qiang Dong, chief scientist of Beijing YuZhi Language Understanding Technology Co. Ltd. and learned more about their YuZhi NLU Platform that conducts its unique semantic analysis based on concepts rather than words.
Andrew Ng, one of the top minds in deep learning, loves teaching. And so, the Stanford University professor decided toContinue Reading
Although food packaging comes with nutrition labels, it’s still not convenient for people to reference. Thus, scientists started to use machine learning algorithms in computer vision to help people determine the caloric value in the food they eat.
PixelGAN is an autoencoder for which the generative path is a convolutional autoregressive neural network on pixels, conditioned on a latent code, and the recognition path uses a generative adversarial network (GAN) to impose a prior distribution on the latent code.
The rubber on GelSight sensor will conform to any object pressed against it, resulting in the reflective lights on the metallic paint to vary according to different materials and shapes.
Multimodal convolutional neural networks for appearance-based gaze estimation that significantly outperforms state-of-the-art methods in the most challenging cross-dataset evaluation
Instead of investigating what steps the brain takes to process information, the authors of this paper link brain activity directly to the learning process.
This paper systematically studies how a convolutional neural network, trained on ImageNet for image classification tasks, works on medical images – or more precisely on ultrasound images – for the “Kidney Detection” problem.
Researchers from Microsoft Research Asia and Shanghai Jiaotong University have recently published a paper on transferring visual attributes of images using a new technique called Deep Image Analogy.
RECOS model supply us with a signal analysis perspective of Convolution Neural Network. From this perspective, we can see the effectiveness of activation and deep architectures.
This paper proposed a computationally efficient convolutional layer to upscale the final low-resolution feature map to a high-resolution output.
Noë proposes his ideas towards current neuroscience research using reductionism approach is far from understanding the philosophy behind our consciousness.