It was a version of a talk that Kearns had given before. But he couldn't ignore the irony of discussing the dangers inherent in new technologies in this particular place. The Santa Fe Institute is just 40 miles from the town of Los Alamos, site of the Manhattan Project, where more than 6,000 scientists and support staff worked together from 1939 to 1945 to produce the world's first atomic bomb. The ultimate impact of the project was enormous: Some 200,000 lives lost at Hiroshima and Nagasaki, and the unleashing of a new technological threat that has loomed over humankind for more than seven decades since. Looking back at those physicists involved in the Manhattan Project and their response to the social and ethical challenges their work presented offers a valuable precedent.

False Data Injection Attacks in Internet of Things and Deep Learning enabled Predictive Analytics Gautam Raj Mode, Prasad Calyam, Khaza Anuarul Hoque Department of Electrical Engineering & Computer Science University of Missouri, Columbia, MO, USA gmwyc@mail.missouri.edu, Abstract --Industry 4.0 is the latest industrial revolution primarily merging automation with advanced manufacturing to reduce direct human effort and resources. Predictive maintenance (PdM) is an industry 4.0 solution, which facilitates predicting faults in a component or a system powered by state-of-the-art machine learning (ML) algorithms (especially deep learning algorithms) and the Internet-of-Things (IoT) sensors. However, IoT sensors and deep learning (DL) algorithms, both are known for their vulnerabilities to cyber-attacks. In the context of PdM systems, such attacks can have catastrophic consequences as they are hard to detect due to the nature of the attack. T o date, the majority of the published literature focuses on the accuracy of the IoT and DL enabled PdM systems and often ignores the effect of such attacks. In this paper, we demonstrate the effect of IoT sensor attacks (in the form of false data injection attack) on a PdM system. At first, we use three state-of-the-art DL algorithms, specifically, Long Short-T erm Memory (LSTM), Gated Recurrent Unit (GRU), and Convolutional Neural Network (CNN) for predicting the Remaining Useful Life (RUL) of a turbofan engine using NASA's C-MAPSS dataset. Our obtained results show that the GRU-based PdM model outperforms some of the recent literature on RUL prediction using the C-MAPSS dataset. Afterward, we model and apply two different types of false data injection attacks (FDIA), specifically, continuous and interim FDIAs on turbofan engine sensor data and evaluate their impact on CNN, LSTM, and GRU-based PdM systems. Our results demonstrate that attacks on even a small number of IoT sensors can strongly defect the RUL prediction in all cases. However, the GRU-based PdM model performs better in terms of accuracy and FDIA resiliency. Lastly, we perform a study on the GRU-based PdM model using four different GRU networks with different sequence lengths.

We develop an adversarial-reinforcement learning scheme for microswimmers in statistically homogeneous and isotropic turbulent fluid flows, in both two (2D) and three dimensions (3D). We show that this scheme allows microswimmers to find non-trivial paths, which enable them to reach a target on average in less time than a na\"ive microswimmer, which tries, at any instant of time and at a given position in space, to swim in the direction of the target. We use pseudospectral direct numerical simulations (DNSs) of the 2D and 3D (incompressible) Navier-Stokes equations to obtain the turbulent flows. We then introduce passive microswimmers that try to swim along a given direction in these flows; the microswimmwers do not affect the flow, but they are advected by it. Two, non-dimensional, control parameters play important roles in our learning scheme: (a) the ratio $\tilde{V}_s$ of the microswimmer's bare velocity $V_s$ and the root-mean-square (rms) velocity $u_{rms}$ of the turbulent fluid; and (b) the product $\tilde{B}$ of the microswimmer-response time $B$ and the rms vorticity $\omega_{rms}$ of the fluid. We show that, in a substantial part of the $\tilde{V}_s-\tilde{B}$ plane, the average time required for the microswimmers to reach the target, by using our adversarial-learning scheme, eventually reduces below the average time taken by microswimmers that follow the na\"ive strategy.

Recent theoretical work has established connections between over-parametrized neural networks and linearized models governed by he Neural Tangent Kernels (NTKs). NTK theory leads to concrete convergence and generalization results, yet the empirical performance of neural networks are observed to exceed their linearized models, suggesting insufficiency of this theory. Towards closing this gap, we investigate the training of over-parametrized neural networks that are beyond the NTK regime yet still governed by the Taylor expansion of the network. We bring forward the idea of \emph{randomizing} the neural networks, which allows them to escape their NTK and couple with quadratic models. We show that the optimization landscape of randomized two-layer networks are nice and amenable to escaping-saddle algorithms. We prove concrete generalization and expressivity results on these randomized networks, which leads to sample complexity bounds (of learning certain simple functions) that match the NTK and can in addition be better by a dimension factor when mild distributional assumptions are present. We demonstrate that our randomization technique can be generalized systematically beyond the quadratic case, by using it to find networks that are coupled with higher-order terms in their Taylor series.

Support vector machines (SVMs) are successful modeling and prediction tools with a variety of applications. Previous work has demonstrated the superiority of the SVMs in dealing with the high dimensional, low sample size problems. However, the numerical difficulties of the SVMs will become severe with the increase of the sample size. Although there exist many solvers for the SVMs, only few of them are designed by exploiting the special structures of the SVMs. In this paper, we propose a highly efficient sparse semismooth Newton based augmented Lagrangian method for solving a large-scale convex quadratic programming problem with a linear equality constraint and a simple box constraint, which is generated from the dual problems of the SVMs. By leveraging the primal-dual error bound result, the fast local convergence rate of the augmented Lagrangian method can be guaranteed. Furthermore, by exploiting the second-order sparsity of the problem when using the semismooth Newton method, the algorithm can efficiently solve the aforementioned difficult problems. Finally, numerical comparisons demonstrate that the proposed algorithm outperforms the current state-of-the-art solvers for the large-scale SVMs.

This paper provides statistical theory and intuition for Personalized PageRank (PPR), a popular technique that samples a small community from a massive network. We study a setting where the entire network is expensive to thoroughly obtain or maintain, but we can start from a seed node of interest and "crawl" the network to find other nodes through their connections. By crawling the graph in a designed way, the PPR vector can be approximated without querying the entire massive graph, making it an alternative to snowball sampling. Using the Degree-Corrected Stochastic Blockmodel, we study whether the PPR vector can select nodes that belong to the same block as the seed node. We provide a simple and interpretable form for the PPR vector, highlighting its biases towards high degree nodes outside of the target block. We examine a simple adjustment based on node degrees and establish consistency results for PPR clustering that allows for directed graphs. We illustrate the method with the Twitter friendship graph and find that (i) the adjusted and unadjusted PPR techniques are complementary approaches, where the adjustment makes the results particularly localized around the seed node and (ii) the bias adjustment greatly benefits from degree regularization.

In this study, a hybrid method based on an Adaptive Neuro-Fuzzy Inference System (ANFIS) and Particle Swarm Optimization (PSO) for diagnosing Liver disorders (ANFIS-PSO) is introduced. This smart diagnosis method deals with a combination of making an inference system and optimization process which tries to tune the hyper-parameters of ANFIS based on the data-set. The Liver diseases characteristics are taken from the UCI Repository of Machine Learning Databases. The number of these characteristic attributes are 7, and the sample number is 354. The right diagnosis performance of the ANFIS-PSO intelligent medical system for liver disease is evaluated by using classification accuracy, sensitivity and specificity analysis, respectively. According to the experimental results, the performance of ANFIS-PSO can be more considerable than traditional FIS and ANFIS without optimization phase.

We introduce the Mutual Information Machine (MIM), a novel formulation of representation learning, using a joint distribution over the observations and latent state in an encoder/decoder framework. Our key principles are symmetry and mutual information, where symmetry encourages the encoder and decoder to learn different factorizations of the same underlying distribution, and mutual information, to encourage the learning of useful representations for downstream tasks. Our starting point is the symmetric Jensen-Shannon divergence between the encoding and decoding joint distributions, plus a mutual information encouraging regularizer. We show that this can be bounded by a tractable cross entropy loss function between the true model and a parameterized approximation, and relate this to the maximum likelihood framework. We also relate MIM to variational autoencoders (VAEs) and demonstrate that MIM is capable of learning symmetric factorizations, with high mutual information that avoids posterior collapse.

--We present a comparative study of different probabilistic forecasting techniques on the task of predicting the electrical load of secondary substations and cabinets located in a low voltage distribution grid, as well as their aggregated power profile. The methods are evaluated using standard KPIs for deterministic and probabilistic forecasts. We also compare the ability of different hierarchical techniques in improving the bottom level forecasters' performances. Both the raw and cleaned datasets, including meteorological data, are made publicly available to provide a standard benchmark for evaluating forecasting algorithms for demand-side management applications. The increasing monitoring capacity in low voltage (L V) and medium voltage (MV) distribution systems allows operators to gather power measurements from different levels of aggregation within the power grid. For instance, smart meters provide measurements from single households or buildings, dedicated power meters or phasor measurement units from secondary substations, and remote terminal units from primary substations at the interface between distribution and (sub)transmission systems. E.g., in a radial distribution system, the power flow at the grid connection point is, at the net of grid losses, the sum of the downstream elements. In the case of forecasts, however, the forecasted top-level series computed by using the information at that level of aggregation does not necessarily correspond to the sum of the bottom-level forecasts, thus invalidating the principle of hierarchy.

We propose a novel defense against all existing gradient based adversarial attacks on deep neural networks for image classification problems. Our defense is based on a combination of deep neural networks and simple image transformations. While straight forward in implementation, this defense yields a unique security property which we term buffer zones. In this paper, we formalize the concept of buffer zones. We argue that our defense based on buffer zones is secure against state-of-the-art black box attacks. We are able to achieve this security even when the adversary has access to the {\em entire} original training data set and unlimited query access to the defense. We verify our security claims through experimentation using FashionMNIST, CIFAR-10 and CIFAR-100. We demonstrate $<10\%$ attack success rate -- significantly lower than what other well-known defenses offer -- at only a price of a 15-20\% drop in clean accuracy. By using a new intuitive metric we explain why this trade-off offers a significant improvement over prior work.