In the midst of the AI "gold rush," how can you separate the nuggets from the fool's gold? There's no shortage of cautionary tales involving overhyped AI claims. And applying AI technologies to health care, education, and law enforcement mean that getting it wrong can have real consequences for society--not just for investors who bet on the wrong unicorn. So IEEE Spectrum asked experts to share their tips for how to identify AI hype in press releases, news articles, research papers, and IPO filings. "It can be tricky, because I think the people who are out there selling the AI hype--selling this AI snake oil--are getting more sophisticated over time," says Tim Hwang, director of the Harvard-MIT Ethics and Governance of AI Initiative.

Recently, a special type of data poisoning (DP) attack targeting Deep Neural Network (DNN) classifiers, known as a backdoor, was proposed. These attacks do not seek to degrade classification accuracy, but rather to have the classifier learn to classify to a target class whenever the backdoor pattern is present in a test example. Launching backdoor attacks does not require knowledge of the classifier or its training process - it only needs the ability to poison the training set with (a sufficient number of) exemplars containing a sufficiently strong backdoor pattern (labeled with the target class). Here we address post-training detection of backdoor attacks in DNN image classifiers, seldom considered in existing works, wherein the defender does not have access to the poisoned training set, but only to the trained classifier itself, as well as to clean examples from the classification domain. This is an important scenario because a trained classifier may be the basis of e.g. a phone app that will be shared with many users. Detecting backdoors post-training may thus reveal a widespread attack. We propose a purely unsupervised anomaly detection (AD) defense against imperceptible backdoor attacks that: i) detects whether the trained DNN has been backdoor-attacked; ii) infers the source and target classes involved in a detected attack; iii) we even demonstrate it is possible to accurately estimate the backdoor pattern. We test our AD approach, in comparison with alternative defenses, for several backdoor patterns, data sets, and attack settings and demonstrate its favorability. Our defense essentially requires setting a single hyperparameter (the detection threshold), which can e.g. be chosen to fix the system's false positive rate.

One of the most important challenges in the analysis of high-throughput genetic data is the development of efficient computational methods to identify statistically significant Single Nucleotide Polymorphisms (SNPs). Genome-wide association studies (GWAS) use single-locus analysis where each SNP is independently tested for association with phenotypes. The limitation with this approach, however, is its inability to explain genetic variation in complex diseases. Alternative approaches are required to model the intricate relationships between SNPs. Our proposed approach extends GWAS by combining deep learning stacked autoencoders (SAEs) and association rule mining (ARM) to identify epistatic interactions between SNPs. Following traditional GWAS quality control and association analysis, the most significant SNPs are selected and used in the subsequent analysis to investigate epistasis. SAERMA controls the classification results produced in the final fully connected multi-layer feedforward artificial neural network (MLP) by manipulating the interestingness measures, support and confidence, in the rule generation process. The best classification results were achieved with 204 SNPs compressed to 100 units (77% AUC, 77% SE, 68% SP, 53% Gini, logloss=0.58, and MSE=0.20), although it was possible to achieve 73% AUC (77% SE, 63% SP, 45% Gini, logloss=0.62, and MSE=0.21) with 50 hidden units - both supported by close model interpretation.

Due to the popularity of context-awareness in the Internet of Things (IoT) and the recent advanced features in the most popular IoT device, i.e., smartphone, modeling and predicting personalized usage behavior based on relevant contexts can be highly useful in assisting them to carry out daily routines and activities. Usage patterns of different categories smartphone apps such as social networking, communication, entertainment, or daily life services related apps usually vary greatly between individuals. People use these apps differently in different contexts, such as temporal context, spatial context, individual mood and preference, work status, Internet connectivity like Wifi? status, or device related status like phone profile, battery level etc. Thus, we consider individuals' apps usage as a multi-class context-aware problem for personalized modeling and prediction. Random Forest learning is one of the most popular machine learning techniques to build a multi-class prediction model. Therefore, in this paper, we present an effective context-aware smartphone apps prediction model, and name it "AppsPred" using random forest machine learning technique that takes into account optimal number of trees based on such multi-dimensional contexts to build the resultant forest. The effectiveness of this model is examined by conducting experiments on smartphone apps usage datasets collected from individual users. The experimental results show that our AppsPred significantly outperforms other popular machine learning classification approaches like ZeroR, Naive Bayes, Decision Tree, Support Vector Machines, Logistic Regression while predicting smartphone apps in various context-aware test cases.

Standard supervised learning procedures are validated against a test set that is assumed to have come from the same distribution as the training data. However, in many problems, the test data may have come from a different distribution. We consider the case of having many labeled observations from one distribution, $P_1$, and making predictions at unlabeled points that come from $P_2$. We combine the high predictive accuracy of random forests (Breiman, 2001) with an importance sampling scheme, where the splits and predictions of the base-trees are done in a weighted manner, which we call Locally Optimized Random Forests. These weights correspond to a non-parametric estimate of the likelihood ratio between the training and test distributions. To estimate these ratios with an unlabeled test set, we make the covariate shift assumption, where the differences in distribution are only a function of the training distributions (Shimodaira, 2000.) This methodology is motivated by the problem of forecasting power outages during hurricanes. The extreme nature of the most devastating hurricanes means that typical validation set ups will overly favor less extreme storms. Our method provides a data-driven means of adapting a machine learning method to deal with extreme events.

The rapid digital transformation without security considerations has resulted in the rise of global-scale cyberattacks. The first line of defense against these attacks are Network Intrusion Detection Systems (NIDS). Once deployed, however, these systems work as blackboxes with a high rate of false positives with no measurable effectiveness. There is a need to continuously test and improve these systems by emulating real-world network attack mutations. We present SynGAN, a framework that generates adversarial network attacks using the Generative Adver-sial Networks (GAN). SynGAN generates malicious packet flow mutations using real attack traffic, which can improve NIDS attack detection rates. As a first step, we compare two public datasets, NSL-KDD and CI-CIDS2017, for generating synthetic Distributed Denial of Service (DDoS) network attacks. We evaluate the attack quality (real vs. synthetic) using a gradient boosting classifier.

Ecologists have long suspected that species are more likely to interact if their traits match in a particular way. For example, a pollination interaction may be particularly likely if the proportions of a bee's tongue match flower shape in a beneficial way. Empirical evidence for trait matching, however, varies significantly in strength among different types of ecological networks. Here, we show that ambiguity among empirical trait matching studies may have arisen at least in parts from using overly simple statistical models. Using simulated and real data, we contrast conventional regression models with Machine Learning (ML) models (Random Forest, Boosted Regression Trees, Deep Neural Networks, Convolutional Neural Networks, Support Vector Machines, naive Bayes, and k-Nearest-Neighbor), testing their ability to predict species interactions based on traits, and infer trait combinations causally responsible for species interactions. We find that the best ML models can successfully predict species interactions in plant-pollinator networks (up to 0.93 AUC) and outperform conventional regression models. Our results also demonstrate that ML models can better identify the causally responsible trait matching combinations than GLMs. In two case studies, the best ML models could successfully predict species interactions in a global plant-pollinator database and infer ecologically plausible trait matching rules for a plant-hummingbird network from Costa Rica, without any prior assumptions about the system. We conclude that flexible ML models offer many advantages over traditional regression models for understanding interaction networks. We anticipate that these results extrapolate to other network types, such as trophic or competitive networks. More generally, our results highlight the potential of ML and artificial intelligence for inference beyond standard tasks such as pattern recognition.

Hotel cancellations can cause issues for many businesses in the industry. Not only is there the lost revenue as a result of the customer canceling, but this can also cause difficulty in coordinating bookings and adjusting revenue management practices. Data analytics can help to overcome this issue, in terms of identifying the customers who are most likely to cancel – allowing a hotel chain to adjust its marketing strategy accordingly. To investigate how machine learning can aid in this task, the ExtraTreesClassifer, logistic regression, and support vector machine models were employed in Python to determine whether cancellations can be accurately predicted with this model. For this example, both hotels are based in Portugal.

Language identification ("LI") is the problem of determining the natural language that a document or part thereof is written in. Automatic LI has been extensively researched for over fifty years. Today, LI is a key part of many text processing pipelines, as text processing techniques generally assume that the language of the input text is known. Research in this area has recently been especially active. This article provides a brief history of LI research, and an extensive survey of the features and methods used in the LI literature. We describe the features and methods using a unified notation, to make the relationships between methods clearer. We discuss evaluation methods, applications of LI, as well as off-the-shelf LI systems that do not require training by the end user. Finally, we identify open issues, survey the work to date on each issue, and propose future directions for research in LI.