Machine learning tools can be useful for historians to analyse large volumes of data and minimize noise, suggests a new study. How do we know if an event is historic? An event's historical significance depends on how it affects subsequent events in the future. But predicting this can be difficult: what may seem historic now may be deemed trivial by future generations. New research suggests that, even with machine learning tools, determining historical significance is difficult but these tools can still help historians.

Risk-related AI applications (risk management, lending, compliance, fraud and cybersecurity) account for 72% of the total $2.8 billion in funds raised for AI vendor companies in banking, according to the latest report by Emerj Artificial Intelligence Research. Compliance and fraud-related applications make up 32% of the total AI vendor landscape in banking, but banks report these applications as a mere 19% of their current AI initiatives. Bankers today see AI as a risk-reduction technology. Their AI initiatives are likely to yield negative ROI in part because they hold naive views about AI's integration and data requirements. Banks are eager to automate compliance specifically, especially given recent data privacy laws such as GDPR.

Imagine a scenario where you need your car's onboard navigation system to place an emergency call, but it won't. Or arriving extra early for every international flight because airport security scanners never recognize your face. For many people--especially people of color and women--these scenarios can be a frustrating reality. That's because the AI that's supposed to make life easier for us all doesn't include diverse enough data to work for everyone. This is a big problem, but one that can be fixed.

Machine learning (ML) and artificial intelligence (AI) algorithms are now being used to automate the discovery of physics principles and governing equations from measurement data alone. However, positing a universal physical law from data is challenging without simultaneously proposing an accompanying discrepancy model to account for the inevitable mismatch between theory and measurements. By revisiting the classic problem of modeling falling objects of different size and mass, we highlight a number of subtle and nuanced issues that must be addressed by modern data-driven methods for the automated discovery of physics. Specifically, we show that measurement noise and complex secondary physical mechanisms, such as unsteady fluid drag forces, can obscure the underlying law of gravitation, leading to an erroneous model. Without proposing an appropriate discrepancy model to handle these drag forces, the data supports an Aristotelian, versus a Galilean, theory of gravitation. Using the sparse identification of nonlinear dynamics (SINDy) algorithm, with the additional assumption that each separate falling object is governed by the same physical law, we are able to identify a viable discrepancy model to account for the fluid dynamic forces that explain the mismatch between a posited universal law of gravity and the measurement data. This work highlights the fact that the simple application of ML/AI will generally be insufficient to extract universal physical laws without further modification.

This paper provides a comprehensive survey of Machine Learning Testing (ML testing) research. It covers 128 papers on testing properties (e.g., correctness, robustness, and fairness), testing components (e.g., the data, learning program, and framework), testing workflow (e.g., test generation and test evaluation), and application scenarios (e.g., autonomous driving, machine translation). The paper also analyses trends concerning datasets, research trends, and research focus, concluding with research challenges and promising research directions in ML testing.

It has been shown that deep neural networks (DNNs) may be vulnerable to adversarial attacks, raising the concern on their robustness particularly for safety-critical applications. Recognizing the local nature and limitations of existing adversarial attacks, we present a new type of global adversarial attacks for assessing global DNN robustness. More specifically, we propose a novel concept of global adversarial example pairs in which each pair of two examples are close to each other but have different class labels predicted by the DNN. We further propose two families of global attack methods and show that our methods are able to generate diverse and intriguing adversarial example pairs at locations far from the training or testing data. Moreover, we demonstrate that DNNs hardened using the strong projected gradient descent (PGD) based (local) adversarial training are vulnerable to the proposed global adversarial example pairs, suggesting that global robustness must be considered while training robust deep learning networks.

The power of neural networks lies in their ability to generalize to unseen data, yet the underlying reasons for this phenomenon remain elusive. Numerous rigorous attempts have been made to explain generalization, but available bounds are still quite loose, and analysis does not always lead to true understanding. The goal of this work is to make generalization more intuitive. Using visualization methods, we discuss the mystery of generalization, the geometry of loss landscapes, and how the curse (or, rather, the blessing) of dimensionality causes optimizers to settle into minima that generalize well.

Elections and opinion polls often have many candidates, with the aim to either rank the candidates or identify a small set of winners according to voters' preferences. In practice, voters do not provide a full ranking; instead, each voter provides their favorite K candidates, potentially in ranked order. The election organizer must choose K and an aggregation rule. We provide a theoretical framework to make these choices. Each K-Approval or K-partial ranking mechanism (with a corresponding positional scoring rule) induces a learning rate for the speed at which the election correctly recovers the asymptotic outcome. Given the voter choice distribution, the election planner can thus identify the rate optimal mechanism. Earlier work in this area provides coarse order-of-magnitude guaranties which are not sufficient to make such choices. Our framework further resolves questions of when randomizing between multiple mechanisms may improve learning, for arbitrary voter noise models. Finally, we use data from 5 large participatory budgeting elections that we organized across several US cities, along with other ranking data, to demonstrate the utility of our methods. In particular, we find that historically such elections have set K too low and that picking the right mechanism can be the difference between identifying the ultimate winner with only a 80% probability or a 99.9% probability after 400 voters.

NASA's Osiris-REx spacecraft has made its closest approach yet to an asteroid 1.4 billion miles from Earth. The probe dipped down to just .4 According to NASA, this put it in position to break the record for'the closest distance a spacecraft has orbited a body in the solar system.' Stunning new images from after the maneuver now reveal a close look at the boulders and craters dotting the surface of the distant object. The close-up image shared this week by NASA was captured on the 13th, showing Bennu half sunlit and half in shadow from the spacecraft's view. At the time, Osiris-REx was .4

What do Russian trolls, Facebook, and US elections have to do with machine learning? Recommendation engines are at the heart of the central feedback loop of social networks and the user-generated content (UGC) they create. Users join the network and are recommended users and content with which to engage. Recommendation engines can be gamed because they amplify the effects of thought bubbles. The 2016 US presidential election showed how important it is to understand how recommendation engines work and the limitations and strengths they offer.