Anomaly detectors are often used to produce a ranked list of statistical anomalies, which are examined by human analysts in order to extract the actual anomalies of interest. Unfortunately, in realworld applications, this process can be exceedingly difficult for the analyst since a large fraction of high-ranking anomalies are false positives and not interesting from the application perspective. In this paper, we aim to make the analyst's job easier by allowing for analyst feedback during the investigation process. Ideally, the feedback influences the ranking of the anomaly detector in a way that reduces the number of false positives that must be examined before discovering the anomalies of interest. In particular, we introduce a novel technique for incorporating simple binary feedback into tree-based anomaly detectors. We focus on the Isolation Forest algorithm as a representative tree-based anomaly detector, and show that we can significantly improve its performance by incorporating feedback, when compared with the baseline algorithm that does not incorporate feedback. Our technique is simple and scales well as the size of the data increases, which makes it suitable for interactive discovery of anomalies in large datasets.

We propose a DTCWT ScatterNet Convolutional Neural Network (DTSCNN) formed by replacing the first few layers of a CNN network with a parametric log based DTCWT ScatterNet. The ScatterNet extracts edge based invariant representations that are used by the later layers of the CNN to learn high-level features. This improves the training of the network as the later layers can learn more complex patterns from the start of learning because the edge representations are already present. The efficient learning of the DTSCNN network is demonstrated on CIFAR-10 and Caltech-101 datasets. The generic nature of the ScatterNet front-end is shown by an equivalent performance to pre-trained CNN front-ends. A comparison with the state-of-the-art on CIFAR-10 and Caltech-101 datasets is also presented.

The data torrent unleashed by current and upcoming instruments requires scalable analysis methods. Machine Learning approaches scale well. However, separating the instrument measurement from the physical effects of interest, dealing with variable errors, and deriving parameter uncertainties is usually an afterthought. Classic forward-folding analyses with Markov Chain Monte Carlo or Nested Sampling enable parameter estimation and model comparison, even for complex and slow-to-evaluate physical models. However, these approaches require independent runs for each data set, implying an unfeasible number of model evaluations in the Big Data regime. Here we present a new algorithm, collaborative nested sampling, for deriving parameter probability distributions for each observation. Importantly, in our method the number of physical model evaluations scales sub-linearly with the number of data sets, and we make no assumptions about homogeneous errors, Gaussianity, the form of the model or heterogeneity/completeness of the observations. Collaborative nested sampling has immediate application in speeding up analyses of large surveys, integral-field-unit observations, and Monte Carlo simulations.

Markov chains and diffusion processes are indispensable tools in machine learning and statistics that are used for inference, sampling, and modeling. With the growth of large-scale datasets, the computational cost associated with simulating these stochastic processes can be considerable, and many algorithms have been proposed to approximate the underlying Markov chain or diffusion. A fundamental question is how the computational savings trade off against the statistical error incurred due to approximations. This paper develops general results that address this question. We bound the Wasserstein distance between the equilibrium distributions of two diffusions as a function of their mixing rates and the deviation in their drifts. We show that this error bound is tight in simple Gaussian settings. Our general result on continuous diffusions can be discretized to provide insights into the computational-statistical trade-off of Markov chains. As an illustration, we apply our framework to derive finite-sample error bounds of approximate unadjusted Langevin dynamics. We characterize computation-constrained settings where, by using fast-to-compute approximate gradients in the Langevin dynamics, we obtain more accurate samples compared to using the exact gradients. Finally, as an additional application of our approach, we quantify the accuracy of approximate zig-zag sampling. Our theoretical analyses are supported by simulation experiments.

A latent space model for a family of random graphs assigns real-valued vectors to nodes of the graph such that edge probabilities are determined by latent positions. Latent space models provide a natural statistical framework for graph visualizing and clustering. A latent space model of particular interest is the Random Dot Product Graph (RDPG), which can be fit using an efficient spectral method; however, this method is based on a heuristic that can fail, even in simple cases. Here, we consider a closely related latent space model, the Logistic RDPG, which uses a logistic link function to map from latent positions to edge likelihoods. Over this model, we show that asymptotically exact maximum likelihood inference of latent position vectors can be achieved using an efficient spectral method. Our method involves computing top eigenvectors of a normalized adjacency matrix and scaling eigenvectors using a regression step. The novel regression scaling step is an essential part of the proposed method. In simulations, we show that our proposed method is more accurate and more robust than common practices. We also show the effectiveness of our approach over standard real networks of the karate club and political blogs.

As more companies begin to experiment with and deploy machine learning in different settings, it's good to look ahead at what future systems might look like. Today, the typical sequence is to gather data, learn some underlying structure, and deploy an algorithm that systematically captures what you've learned. Gathering, preparing, and enriching the right data--particularly training data--is essential and remains a key bottleneck among companies wanting to use machine learning. I take for granted that future AI systems will rely on continuous learning as opposed to algorithms that are trained offline. Humans learn this way, and AI systems will increasingly have the capacity to do the same.

Hi! I'm Jose Portilla and I'm an instructor on Udemy with over 250,000 students enrolled across various courses on Python for Data Science and Machine Learning, R Programming for Data Science, Python for Big Data, and many more. What should I do to become a data scientist? In this post, I'll try my best to help answer this question and point to resources that can help guide you to an answer, also hopefully this post serves as something I can quickly link to my students:) I've broken down the steps into some key topics and discussed helpful details for each. "The secret of getting ahead is getting started." If you are interested in becoming a data scientist the best advice is to begin preparing for your journey now!

With hype around artificial intelligence (AI) at an all-time high, Gartner says many CIOs are understandably cautious about promoting its potential value to their business. But a new report from Gartner calls on CIOs to start investing now on AI. Digital business pressure combined with the rapid pace of innovation make it a great time for CIOs to aggressively learn how AI might influence the business strategy over the next two to four years, according to the authors of the report, Gartner analysts Janelle B. Hill, Bern Elliot and Jamie Popkin. They say CIOs can get involved at the front end of business strategy development, and educating their CEO and the board about recent developments in AI. In these discussions, CIOs can promote AI's massive potential to disrupt markets and remake existing business models, not just as an output that further automates existing capabilities.

The original Amazon Echo in black is currently out of stock on the company's site, as rumors claim the online retailer will come out with a new version to compete with Apple's HomePod. Amazon has released multiple Alexa-powered Echo devices, including the Echo Look and Echo Show, which were both launched this year. However, the original Echo hasn't been refreshed in two years, and has seen its price slashed in half for Amazon's Prime Day sale. The recent unavailability and discounts on the device suggests Amazon is gearing up to release a new Echo to rival the upcoming HomePod. Apple introduced the HomePod in June, and is expected to launch the speaker in December. Amazon Echo out of stock as rumors point to new device that will rival Apple's HomePod.

When I was studying political science in college, I had no intention of going into the field of technology. I had friends in STEM, but I was sure I either wanted to pursue a career in politics or business. However, when I saw an opportunity to enter a rotational program at IBM Watson starting in the summer of 2014, I knew I had to pursue it. I got the job and rotated through the sales and marketing departments, where I began learning more about AI technology. As I talked to developers both inside and outside of the company, I found myself wanting to learn how to code with the Watson API's and create a new product or app.