Graph convolutional networks are becoming indispensable for deep learning from graph-structured data. Most of the existing graph convolutional networks share two big shortcomings. First, they are essentially low-pass filters, thus the potentially useful middle and high frequency band of graph signals are ignored. Second, the bandwidth of existing graph convolutional filters is fixed. Parameters of a graph convolutional filter only transform the graph inputs without changing the curvature of a graph convolutional filter function. In reality, we are uncertain about whether we should retain or cut off the frequency at a certain point unless we have expert domain knowledge. In this paper, we propose Automatic Graph Convolutional Networks (AutoGCN) to capture the full spectrum of graph signals and automatically update the bandwidth of graph convolutional filters. While it is based on graph spectral theory, our AutoGCN is also localized in space and has a spatial form. Experimental results show that AutoGCN achieves significant improvement over baseline methods which only work as low-pass filters.

In this post, you learn how to create a MLOps project to automate the deployment of an Amazon SageMaker endpoint with multiple production variants for A/B testing. You also deploy a general purpose API and testing infrastructure that includes a multi-armed bandit experiment framework. This testing infrastructure will automatically optimize traffic to the best-performing model over time based on user feedback. Amazon SageMaker MLOps projects are a new capability recently released with Amazon SageMaker Pipelines, the first purpose-built, easy-to-use, continuous integration and continuous delivery (CI/CD) service for ML. The MLOps project template provisions the initial setup required for a complete end-to-end MLOps system, including model building, training, and deployment, and can be customized to support your own organizations requirements.

Retired Navy SEAL Commander Dave Sears suggests Russia, China and Pakistan could face national security issues once U.S. troops leave Afghanistan. This is a rush transcript of "Your World with Neil Cavuto" on July 8, 2021. This copy may not be in its final form and may be updated. QUESTION: Do you trust the Taliban, Mr. President? Do you trust the Taliban, sir? JOE BIDEN, PRESIDENT OF THE UNITED STATES: Are you -- is that a serious question? QUESTION: It is absolutely a serious question. Do you trust the Taliban? BIDEN: No, I do not. BIDEN: No, I do not trust the Taliban. QUESTION: Is the U.S. responsible for the deaths that happen the Afghans after you leave the country? QUESTION: Mr. President, will you amplify that question, please? Will you amplify your answer, please, why you don't trust the Taliban? BIDEN: It is a silly question. Do I trust the Taliban? And it almost seemed like a Donald Trump press conference, with angry reporters trying to get a simple answer from the president, and their agitation showing, as the questions and the nonanswers went on, all of this at a time U.S. forces are moving rapidly ahead of schedule. Better than 90 percent now have left Afghanistan. And we could see them all out well before the 9/11 deadline that the president has set. But he says he's not going to change his mind. And he says that, after 20 years, Afghans must look after themselves. Jennifer Griffin has more from the Pentagon.

You don't want your face to appear in the database of Clearview AI? The company's CEO doesn't seem to care. "All the information we collect is collected legally and it is all publicly available information," Hoan Ton-That said Monday during DW's Global Media Forum (GMF), addressing criticism that the firm's controversial technology infringes on the privacy of hundreds of millions. Privacy activists recently lodged data protection complaints against Clearview AI in five European countries. They argue that the software -- a search engine for faces combing through billions of photos -- violates the UK's and the EU's strict privacy rules.

Registrations are now open for the 5th World of Drones and Robotics Congress (WoDaRC), which will be held at the Brisbane Convention & Exhibition Centre, 18 -19 August 2021. WoDaRC will again be presented as a live physical congress for those who can attend with virtual options for those who cannot. Exhibitors now have the option of lower-cost display "pods" in a new exhibition area which will include networking spaces and food service. Virtual exhibits will also be available. Attend, network, speak, exhibit or watch WoDaRC in the manner that best suits you.

Deep Neural Networks (DNN) are known to be vulnerable to adversarial samples, the detection of which is crucial for the wide application of these DNN models. Recently, a number of deep testing methods in software engineering were proposed to find the vulnerability of DNN systems, and one of them, i.e., Model Mutation Testing (MMT), was used to successfully detect various adversarial samples generated by different kinds of adversarial attacks. However, the mutated models in MMT are always huge in number (e.g., over 100 models) and lack diversity (e.g., can be easily circumvented by high-confidence adversarial samples), which makes it less efficient in real applications and less effective in detecting high-confidence adversarial samples. In this study, we propose Graph-Guided Testing (GGT) for adversarial sample detection to overcome these aforementioned challenges. GGT generates pruned models with the guide of graph characteristics, each of them has only about 5% parameters of the mutated model in MMT, and graph guided models have higher diversity. The experiments on CIFAR10 and SVHN validate that GGT performs much better than MMT with respect to both effectiveness and efficiency.

For machine learning systems to be reliable, we must understand their performance in unseen, out-of-distribution environments. In this paper, we empirically show that out-of-distribution performance is strongly correlated with in-distribution performance for a wide range of models and distribution shifts. Specifically, we demonstrate strong correlations between in-distribution and out-of-distribution performance on variants of CIFAR-10 & ImageNet, a synthetic pose estimation task derived from YCB objects, satellite imagery classification in FMoW-WILDS, and wildlife classification in iWildCam-WILDS. The strong correlations hold across model architectures, hyperparameters, training set size, and training duration, and are more precise than what is expected from existing domain adaptation theory. To complete the picture, we also investigate cases where the correlation is weaker, for instance some synthetic distribution shifts from CIFAR-10-C and the tissue classification dataset Camelyon17-WILDS. Finally, we provide a candidate theory based on a Gaussian data model that shows how changes in the data covariance arising from distribution shift can affect the observed correlations.

Continual learning-the ability to learn many tasks in sequence-is critical for artificial learning systems. Yet standard training methods for deep networks often suffer from catastrophic forgetting, where learning new tasks erases knowledge of earlier tasks. While catastrophic forgetting labels the problem, the theoretical reasons for interference between tasks remain unclear. Here, we attempt to narrow this gap between theory and practice by studying continual learning in the teacher-student setup. We extend previous analytical work on two-layer networks in the teacher-student setup to multiple teachers. Using each teacher to represent a different task, we investigate how the relationship between teachers affects the amount of forgetting and transfer exhibited by the student when the task switches. In line with recent work, we find that when tasks depend on similar features, intermediate task similarity leads to greatest forgetting. However, feature similarity is only one way in which tasks may be related. The teacher-student approach allows us to disentangle task similarity at the level of readouts (hidden-to-output weights) and features (input-to-hidden weights). We find a complex interplay between both types of similarity, initial transfer/forgetting rates, maximum transfer/forgetting, and long-term transfer/forgetting. Together, these results help illuminate the diverse factors contributing to catastrophic forgetting.

The Fisher information is one of the most fundamental concepts in statistical machine learning. Intuitively, it measures the amount of information carried by a single random observation when the underlying model varies along certain directions in the parameter space: if such a variation does not change the underlying model, then a corresponding observation contains zero (Fisher) information and is non-informative regarding the varied parameter. Parameter estimation is impossible in this case. Otherwise, if the variation significantly changes the model and has large information, then an observation is informative and the parameter estimation can be more efficient as compared to parameters with small Fisher information. In machine learning, this basic concept is useful for defining intrinsic structures of the parameter space, measuring model complexity, and performing gradient-based optimization.

"Uncle Stock is one of the best stock screeners up there and I have tested a few. Advising what to do next is a must for my stock market investment. "I would like to thank you immensely for creating such a marvellous gem which contains extensive details preventing us from undergoing the pain to calculate them. This is simply mind blowing. I have everything that I would ever need with tremendous customisation options.