Are you a junior researcher with the potential to become a world-class machine learning scientist? Apply to become a Vector Institute Postdoctoral Fellow and conduct cutting-edge fundamental research in machine learning and deep learning algorithms and their applications. Postdoctoral fellows at the Vector Institute are junior researchers with the potential to become world-class researchers. Like postdoctoral researchers in a University lab, postdoctoral fellows at the Vector Institute are tasked with and supported in carrying out state-of-the-art research, publishing at the highest international level, and contributing to the academic life and reputation of the Institute. In addition, postdoctoral fellows at the Vector Institute have access to the resources of a well-funded institute dedicated solely to machine learning and deep learning, and are encouraged to work with any of our over 25 world-class faculty in machine learning and deep learning, though they will typically work primarily with 1–2 faculty members.

The Omnitek deep learning processing unit (DPU) employs a novel mathematical framework combining low-precision fixed point maths with floating point maths to achieve 135 GOPS/W at full 32-bit floating point accuracy when running the VGG-16 CNN in an Arria 10 GX 1150. Scalable across a wide range of Arria 10 GX and Stratix 10 GX devices, the DPU can be tuned for low cost or high performance in either embedded or data centre applications. The DPU is fully software programmable in C/C or Python using standard frameworks such as TensorFlow, enabling it to be configured for a wide range of standard CNN models including GoogLeNet, ResNet-50 and VGG-16 as well as custom models. No FPGA design expertise is required to do this. "We are very excited to apply this unique innovation, resulting from our joint research program with Oxford University, to reducing the cost of a whole slew of AI-enabled applications, particularly in video and imaging where we have a rich library of highly optimised IP to complement the DPU and create complete systems on a chip", commented Roger Fawcett, CEO at Omnitek.

Whenever we start to talk about artificial intelligence, machine learning, or deep learning, the cautionary tales from science fiction cinema arise: HAL 9000 from 2001: A Space Odyssey, the T-series robots from Terminator, replicants from Blade Runner, there are hundreds of stories about computers learning too much and becoming a threat. The crux of these movies always has one thing in common: there are things that computers do well and things that humans can do well, and they don't necessarily intersect. Computers are really good at crunching numbers and statistical analysis (deductive reasoning) and humans are really good at recognizing patterns and making inductive decisions using deductive data. Both have their strengths and their role. With the massive proliferation of data across platforms, types, and collection schedules, how are geospatial specialists supposed to address this apparently insurmountable task?

Go player Lee Sedol (R) during the third game of the Google DeepMind Challenge Match against Google-developed supercomputer AlphaGo. Leading Australian artificial intelligence scientist Professor Toby Walsh is warning that we are "sleepwalking" into an AI future in which billions of machines and computers will be able to think. Professor Walsh, from the University of New South Wales, is calling for a national discussion about whether society needs to adopt clear boundaries and guidelines around how AI is developed and how it's used in our lives. In his book It's Alive: Artificial Intelligence From The Logic Piano to Killer Robots, he has highlighted key questions in a series of predictions that describe how our future could be far better or far worse because of AI. Here's how he thinks society might change by 2050 thanks to artificial intelligence.

How to model and encode the semantics of human-written text and select the type of neural network to process it are not settled issues in sentiment analysis. Accuracy and transferability are critical issues in machine learning in general. These properties are closely related to the loss estimates for the trained model. I present a computationally-efficient and accurate feedforward neural network for sentiment prediction capable of maintaining low losses. When coupled with an effective semantics model of the text, it provides highly accurate models with low losses. Experimental results on representative benchmark datasets and comparisons to other methods show the advantages of the new approach.

Using large mini-batches when training generative adversarial networks (GANs) has been recently shown to significantly improve the quality of the generated samples. This can be seen as a simple but computationally expensive way of reducing the noise of the gradient estimates. In this paper, we investigate the effect of the noise in this context and show that it can prevent the convergence of standard stochastic game optimization methods, while their respective batch version converges. To address this issue, we propose a variance-reduced version of the stochastic extragradient algorithm (SVRE). We show experimentally that it performs similarly to a batch method, while being computationally cheaper, and show its theoretical convergence, improving upon the best rates proposed in the literature. Experiments on several datasets show that SVRE improves over baselines. Notably, SVRE is the first optimization method for GANs to our knowledge that can produce near state-of-the-art results without using adaptive step-size such as Adam.

Rumours have existed for a long time and have been known for serious consequences. The rapid growth of social media platforms has multiplied the negative impact of rumours; it thus becomes important to early detect them. Many methods have been introduced to detect rumours using the content or the social context of news. However, most existing methods ignore or do not explore effectively the propagation pattern of news in social media, including the sequence of interactions of social media users with news across time. In this work, we propose a novel method for rumour detection based on deep learning. Our method leverages the propagation process of the news by learning the users' representation and the temporal interrelation of users' responses. Experiments conducted on Twitter and Weibo datasets demonstrate the state-of-the-art performance of the proposed method.

Body-worn cameras are now commonly used for logging daily life, sports, and law enforcement activities, creating a large volume of archived footage. This paper studies the problem of classifying frames of footage according to the activity of the camera-wearer with an emphasis on application to real-world police body-worn video. Real-world datasets pose a different set of challenges from existing egocentric vision datasets: the amount of footage of different activities is unbalanced, the data contains personally identifiable information, and in practice it is difficult to provide substantial training footage for a supervised approach. We address these challenges by extracting features based exclusively on motion information then segmenting the video footage using a semi-supervised classification algorithm. On publicly available datasets, our method achieves results comparable to, if not better than, supervised and/or deep learning methods using a fraction of the training data. It also shows promising results on real-world police body-worn video.

ProductNet is a collection of high-quality product datasets for better product understanding. Motivated by ImageNet, ProductNet aims at supporting product representation learning by curating product datasets of high quality with properly chosen taxonomy. In this paper, the two goals of building high-quality product datasets and learning product representation support each other in an iterative fashion: the product embedding is obtained via a multi-modal deep neural network (master model) designed to leverage product image and catalog information; and in return, the embedding is utilized via active learning (local model) to vastly accelerate the annotation process. For the labeled data, the proposed master model yields high categorization accuracy (94.7% top-1 accuracy for 1240 classes), which can be used as search indices, partition keys, and input features for machine learning models. The product embedding, as well as the fined-tuned master model for a specific business task, can also be used for various transfer learning tasks.

Traditional applications of GNNs assume an a priori We explore the use of graph neural networks notion of entity (such as bodies, links or particles) (GNNs) to model spatial processes in which and match every node in the graph to an entity. We there is no a priori graphical structure. Similar propose to apply GNNs to the problem of modeling to finite element analysis, we assign nodes transformations of functions defined on continuous of a GNN to spatial locations and use a computational spaces, using a structure we call graph element networks process defined on the graph to (GENs). Inspired by finite element methods, we use model the relationship between an initial graph neural networks to mesh a continuous space function defined over a space and a resulting and define an iterative computation that propagates function in the same space. We use GNNs information from some sampled input values in the as a computational substrate, and show that space to an output function defined everywhere in the locations of the nodes in space as well the space. GENs allow us to model systems that have as their connectivity can be optimized to focus spatial structure but lack a clear notion of entity, such on the most complex parts of the space.