An important issue in clustering concerns the avoidance of false positives while searching for clusters. This work addressed this problem considering agglomerative methods, namely single, average, median, complete, centroid and Ward's approaches applied to unimodal and bimodal datasets obeying uniform, gaussian, exponential and power-law distributions. A model of clusters was also adopted, involving a higher density nucleus surrounded by a transition, followed by outliers. This paved the way to defining an objective means for identifying the clusters from dendrograms. The adopted model also allowed the relevance of the clusters to be quantified in terms of the height of their subtrees. The obtained results include the verification that many methods detect two clusters in unimodal data. The single-linkage method was found to be more resilient to false positives. Also, several methods detected clusters not corresponding directly to the nucleus. The possibility of identifying the type of distribution was also investigated.

Continually learning new skills is important for intelligent systems, yet standard deep learning methods suffer from catastrophic forgetting of the past. Recent works address this with weight regularisation. Functional regularisation, although computationally expensive, is expected to perform better, but rarely does so in practice. In this paper, we fix this issue by using a new functional-regularisation approach that utilises a few memorable past examples crucial to avoid forgetting. By using a Gaussian Process formulation of deep networks, our approach enables training in weight-space while identifying both the memorable past and a functional prior. Our method achieves state-of-the-art performance on standard benchmarks and opens a new direction for life-long learning where regularisation and memory-based methods are naturally combined.

Deep learning is widely known for its flexibility and the capability to uncover complex patterns in large datasets; with these advantages, instances of deep learning in the genomics domain are emerging. One such application is genomic prediction, where the traits of individuals -- like susceptibility to disease or yield-related traits -- are predicted using their genomic information. Understanding the correlation of the genetic traits and variations in genomes could have many benefits such as advancing crop breeding processes, and hence improve food security. In this article, we explore how genetic relationships can be exploited alongside genomic information to predict genetic traits, with the aid of graph machine learning algorithms. In genomic prediction, traditional deep learning would use an individual's genomic information -- like a single nucleotide polymorphism (SNP) -- as input features to the neural network. A SNP is essentially a difference that occurs at a specific position in an individual's genome.

A diver examines one of the coral nurseries on the coral reefs of the Society Islands in French ... [ ] Polynesia. While scientists have succeeded at restoring some coral reefs, humans alone can't save all the reefs that are dying across the globe, a NOAA reef restoration manager said this month. Even in the best of conditions, human divers can spend only three or four hours per day working under water, said Tom Moore, coral reef restoration program manager for the National Oceanic And Atmospheric Administration. And those best conditions are rare. That's not enough to halt the collapse of one of the planet's most crucial ecosystems, Moore said at the EarthxOcean conference: half the world's coral reefs have died and the rest are expected to perish in this century.

Deep learning (DL) techniques have gained significant popularity among software engineering (SE) researchers in recent years. This is because they can often solve many SE challenges without enormous manual feature engineering effort and complex domain knowledge. Although many DL studies have reported substantial advantages over other state-of-the-art models on effectiveness, they often ignore two factors: (1) replicability - whether the reported experimental result can be approximately reproduced in high probability with the same DL model and the same data; and (2) reproducibility - whether one reported experimental findings can be reproduced by new experiments with the same experimental protocol and DL model, but different sampled real-world data. Unlike traditional machine learning (ML) models, DL studies commonly overlook these two factors and declare them as minor threats or leave them for future work. This is mainly due to high model complexity with many manually set parameters and the time-consuming optimization process. In this study, we conducted a literature review on 93 DL studies recently published in twenty SE journals or conferences. Our statistics show the urgency of investigating these two factors in SE. Moreover, we re-ran four representative DL models in SE. Experimental results show the importance of replicability and reproducibility, where the reported performance of a DL model could not be replicated for an unstable optimization process. Reproducibility could be substantially compromised if the model training is not convergent, or if performance is sensitive to the size of vocabulary and testing data. It is therefore urgent for the SE community to provide a long-lasting link to a replication package, enhance DL-based solution stability and convergence, and avoid performance sensitivity on different sampled data.

Influence functions approximate the effect of training samples in test-time predictions and have a wide variety of applications in machine learning interpretability and uncertainty estimation. A commonly-used (first-order) influence function can be implemented efficiently as a post-hoc method requiring access only to the gradients and Hessian of the model. For linear models, influence functions are well-defined due to the convexity of the underlying loss function and are generally accurate even across difficult settings where model changes are fairly large such as estimating group influences. Influence functions, however, are not well-understood in the context of deep learning with non-convex loss functions. In this paper, we provide a comprehensive and large-scale empirical study of successes and failures of influence functions in neural network models trained on datasets such as Iris, MNIST, CIFAR-10 and ImageNet. Through our extensive experiments, we show that the network architecture, its depth and width, as well as the extent of model parameterization and regularization techniques have strong effects in the accuracy of influence functions. In particular, we find that (i) influence estimates are fairly accurate for shallow networks, while for deeper networks the estimates are often erroneous; (ii) for certain network architectures and datasets, training with weight-decay regularization is important to get high-quality influence estimates; and (iii) the accuracy of influence estimates can vary significantly depending on the examined test points. These results suggest that in general influence functions in deep learning are fragile and call for developing improved influence estimation methods to mitigate these issues in non-convex setups.

Forecasting the weather is an increasingly data intensive exercise. Numerical Weather Prediction (NWP) models are becoming more complex, with higher resolutions, and there are increasing numbers of different models in operation. While the forecasting skill of NWP models continues to improve, the number and complexity of these models poses a new challenge for the operational meteorologist: how should the information from all available models, each with their own unique biases and limitations, be combined in order to provide stakeholders with well-calibrated probabilistic forecasts to use in decision making? In this paper, we use a road surface temperature example to demonstrate a three-stage framework that uses machine learning to bridge the gap between sets of separate forecasts from NWP models and the 'ideal' forecast for decision support: probabilities of future weather outcomes. First, we use Quantile Regression Forests to learn the error profile of each numerical model, and use these to apply empirically-derived probability distributions to forecasts. Second, we combine these probabilistic forecasts using quantile averaging. Third, we interpolate between the aggregate quantiles in order to generate a full predictive distribution, which we demonstrate has properties suitable for decision support. Our results suggest that this approach provides an effective and operationally viable framework for the cohesive post-processing of weather forecasts across multiple models and lead times to produce a well-calibrated probabilistic output.

Bayesian network structure learning is the notoriously difficult problem of discovering a Bayesian network (BN) that optimally represents a given set of training data [4]. Since the exact inference on a BN is exponential in the BN's treewidth [14], one is particularly interested in learning BNs of bounded treewidth. However, learning a BN of bounded treewidth that optimally fits the data (i.e., with the largest possible score) is, in turn, an NPhard task [13]. This predicament caused the research on treewidth-bounded BN structure learning to split into two branches: 1. Heuristic Learning (see, e.g., [5, 17, 23, 24]), which is scalable to large BNs with thousands of nodes (but with a score that can be far from optimal), and 2. Exact Learning (see, e.g., [2, 13, 19]), which learns optimal BNs (but is scalable only to a few dozen nodes). In this paper, we combine heuristic and exact learning and take the best of both worlds.

Strong passwords are fundamental to the security of password-based user authentication systems. In recent years, much effort has been made to evaluate password strength or to generate strong passwords. Unfortunately, the usability or memorability of the strong passwords has been largely neglected. In this paper, we aim to bridge the gap between strong password generation and the usability of strong passwords. We propose to automatically generate textual password mnemonics, i.e., natural language sentences, which are intended to help users better memorize passwords. We introduce \textit{DeepMnemonic}, a deep attentive encoder-decoder framework which takes a password as input and then automatically generates a mnemonic sentence for the password. We conduct extensive experiments to evaluate DeepMnemonic on the real-world data sets. The experimental results demonstrate that DeepMnemonic outperforms a well-known baseline for generating semantically meaningful mnemonic sentences. Moreover, the user study further validates that the generated mnemonic sentences by DeepMnemonic are useful in helping users memorize strong passwords.

We develop a method for generating causal post-hoc explanations of black-box classifiers based on a learned low-dimensional representation of the data. The explanation is causal in the sense that changing learned latent factors produces a change in the classifier output statistics. To construct these explanations, we design a learning framework that leverages a generative model and information-theoretic measures of causal influence. Our objective function encourages both the generative model to faithfully represent the data distribution and the latent factors to have a large causal influence on the classifier output. Our method learns both global and local explanations, is compatible with any classifier that admits class probabilities and a gradient, and does not require labeled attributes or knowledge of causal structure. Using carefully controlled test cases, we provide intuition that illuminates the function of our causal objective. We then demonstrate the practical utility of our method on image recognition tasks.