Voyager 1 is still alive out there, barreling into the cosmos more than 15 billion miles away. However, a computer problem has kept the mission's loyal support team in Southern California from knowing much more about the status of one of NASA's longest-lived spacecraft. The computer glitch cropped up on November 14, and it affected Voyager 1's ability to send back telemetry data, such as measurements from the craft's science instruments or basic engineering information about how the probe was doing. As a result, the team has no insight into key parameters regarding the craft's propulsion, power, or control systems. "It would be the biggest miracle if we get it back. We certainly haven't given up," said Suzanne Dodd, Voyager project manager at NASA's Jet Propulsion Laboratory, in an interview with Ars.

You can read this post on our blog. Now let's dive into the activity at and around rOpenSci! Consult our Events page to find your local time and how to join. Find out about more events. Maëlle Salmon (Research Software Engineer with rOpenSci) and Karthik Ram (rOpenSci executive director) authored a commentary "The R Developer Community Does Have a Strong Software Engineering Culture" in the latest issue of The R Journal edited by Di Cook, as a response to the discussion paper "Software Engineering and R Programming: A Call for Research" by Melina Vidoni (who's an Associate editor of rOpenSci Software Peer Review).

In this article, we'll train a PyTorch model to perform super-resolution imaging, a technique for gracefully upscaling images. Super-resolution imaging (right) infers pixel values from a lower-resolution image (left). Machine learning projects typically begin by acquiring data, cleaning the data, and converting the data into model-native formats. Such manual data pipelines are tedious to create and difficult to reproduce over time, across collaborators, and across machines. Moreover, trained models are often stored haphazardly, without version control.

Recently there has been interest in using satellite images as investing tools. Hedge funds are looking at construction in Beijing to bet on concrete demand, or they are counting cars in Walmart parking lots to get an early estimates on profits. Here I discuss a project to determine land use (i.e. is an area a building or not a building) from satellite images. The idea was to measure the change in land use over time as an economic indicator. This project was a proof of concept for the Insight Data Fellows Program.

Think about how much we need cognitive technologies that enable us to make accurate estimates. There is a need for knowledge to save the potato from speculators, to predict elections and the weather accurately, i.e. cognitive computing. Will it hail, when will it hail, how long will it last, how big will the hailstones be and where will it hail in Istanbul from Tekirdağ to Kocaeli? Not Nostradamus but big data and cognitive computing technology lets you find the right answers. Weather forecasting It was not in vain that IBM bought The Weather Company, which has the world's most sensitive, precise and reliable weather data, at the beginning of 2016.

Real world multimedia data is often composed of multiple modalities such as an image or a video with associated text (e.g. captions, user comments, etc.) and metadata. Such multimodal data packages are prone to manipulations, where a subset of these modalities can be altered to misrepresent or repurpose data packages, with possible malicious intent. It is, therefore, important to develop methods to assess or verify the integrity of these multimedia packages. Using computer vision and natural language processing methods to directly compare the image (or video) and the associated caption to verify the integrity of a media package is only possible for a limited set of objects and scenes. In this paper, we present a novel deep learning-based approach for assessing the semantic integrity of multimedia packages containing images and captions, using a reference set of multimedia packages. We construct a joint embedding of images and captions with deep multimodal representation learning on the reference dataset in a framework that also provides image-caption consistency scores (ICCSs). The integrity of query media packages is assessed as the inlierness of the query ICCSs with respect to the reference dataset. We present the MultimodAl Information Manipulation dataset (MAIM), a new dataset of media packages from Flickr, which we make available to the research community. We use both the newly created dataset as well as Flickr30K and MS COCO datasets to quantitatively evaluate our proposed approach. The reference dataset does not contain unmanipulated versions of tampered query packages. Our method is able to achieve F1 scores of 0.75, 0.89 and 0.94 on MAIM, Flickr30K and MS COCO, respectively, for detecting semantically incoherent media packages.

Feature selection is a process of extracting valuable features that have significant influence on dependent variable. This is still an active field of research and machine wandering. In this post I compare few feature selection algorithms: traditional GLM with regularization, computationally demanding Boruta and entropy based filter from FSelectorRcpp (free of Java/Weka) package. Check out the comparison on Venn Diagram carried out on data from the RTCGA factory of R data packages. I would like to thank Magda Sobiczewska and pbiecek for inspiration for this comparison.