During the last years, a remarkable breakthrough has been made in AI domain thanks to artificial deep neural networks that achieved a great success in many machine learning tasks in computer vision, natural language processing, speech recognition, malware detection and so on. However, they are highly vulnerable to easily crafted adversarial examples. Many investigations have pointed out this fact and different approaches have been proposed to generate attacks while adding a limited perturbation to the original data. The most robust known method so far is the so called C&W attack [1]. Nonetheless, a countermeasure known as feature squeezing coupled with ensemble defense showed that most of these attacks can be destroyed [6]. In this paper, we present a new method we call Centered Initial Attack (CIA) whose advantage is twofold : first, it insures by construction the maximum perturbation to be smaller than a threshold fixed beforehand, without the clipping process that degrades the quality of attacks. Second, it is robust against recently introduced defenses such as feature squeezing, JPEG encoding and even against a voting ensemble of defenses. While its application is not limited to images, we illustrate this using five of the current best classifiers on ImageNet dataset among which two are adversarialy retrained on purpose to be robust against attacks. With a fixed maximum perturbation of only 1.5% on any pixel, around 80% of attacks (targeted) fool the voting ensemble defense and nearly 100% when the perturbation is only 6%. While this shows how it is difficult to defend against CIA attacks, the last section of the paper gives some guidelines to limit their impact.

Time series forecasting is ubiquitous in the modern world. Applications range from health care to astronomy, include climate modelling, financial trading and monitoring of critical engineering equipment. To offer value over this range of activities we must have models that not only provide accurate forecasts but that also quantify and adjust their uncertainty over time. Furthermore, such models must allow for multimodal, non-Gaussian behaviour that arises regularly in applied settings. In this work, we propose a novel, end-to-end deep learning method for time series forecasting. Crucially, our model allows the principled assessment of predictive uncertainty as well as providing rich information regarding multiple modes of future data values. Our approach not only provides an excellent predictive forecast, shadowing true future values, but also allows us to infer valuable information, such as the predictive distribution of the occurrence of critical events of interest, accurately and reliably even over long time horizons. We find the method outperforms other state-of-the-art algorithms, such as Gaussian Processes.

The new wave of successful generative models in machine learning has increased the interest in deep learning driven de novo drug design. However, assessing the performance of such generative models is notoriously difficult. Metrics that are typically used to assess the performance of such generative models are the percentage of chemically valid molecules or the similarity to real molecules in terms of particular descriptors, such as the partition coefficient (logP) or druglikeness. However, method comparison is difficult because of the inconsistent use of evaluation metrics, the necessity for multiple metrics, and the fact that some of these measures can easily be tricked by simple rule-based systems. We propose a novel distance measure between two sets of molecules, called Fr\'echet ChemblNet distance (FCD), that can be used as an evaluation metric for generative models. The FCD is similar to a recently established performance metric for comparing image generation methods, the Fr\'echet Inception Distance (FID). Whereas the FID uses one of the hidden layers of InceptionNet, the FCD utilizes the penultimate layer of a deep neural network called "ChemblNet", which was trained to predict drug activities. Thus, the FCD metric takes into account chemically and biologically relevant information about molecules, and also measures the diversity of the set via the distribution of generated molecules. The FCD's advantage over previous metrics is that it can detect if generated molecules are a) diverse and have similar b) chemical and c) biological properties as real molecules. We further provide an easy-to-use implementation that only requires the SMILES representation of the generated molecules as input to calculate the FCD. Implementations are available at: https://www.github.com/bioinf-jku/FCD.

Variational Autoencoder (VAE) is one of the most popular generative models, and enormous advances have been explored in recent years. Due to the increasing complexity of the raw data and the model architecture, deep networks are needed in VAE models while few works discuss their impacts. According to our observation, VAE does not always benefit from deeper architecture: 1) Deeper encoder makes VAE learn more comprehensible latent representations, while results in blurry reconstruction samples; 2) Deeper decoder ensures more high-quality generations, while the latent representations become abstruse; 3) When encoder and decoder both go deeper, abstruse latent representation occurs with blurry reconstruction samples at same time. In this paper, we deduce a Fisher information measure for the corresponding analysis. With such measure, we demonstrate that information loss is ineluctable in feed-forward networks and causes the previous three types of degeneration, especially when the network goes deeper. We also demonstrate that skip connections benefit the preservation of information amount, thus propose a VAE enhanced by skip connections, named SCVAE. In the experiments, SCVAE is shown to mitigate the information loss and to achieve a promising performance in both encoding and decoding tasks. Moreover, SCVAE can be adaptive to other state-of-the-art variants of VAE for further amelioration.

Amortized variational inference (AVI) replaces instance-specific local inference with a global inference network. While AVI has enabled efficient training of deep generative models such as variational autoencoders (VAE), recent empirical work suggests that inference networks can produce suboptimal variational parameters. We propose a hybrid approach, to use AVI to initialize the variational parameters and run stochastic variational inference (SVI) to refine them. Crucially, the local SVI procedure is itself differentiable, so the inference network and generative model can be trained end-to-end with gradient-based optimization. This semi-amortized approach enables the use of rich generative models without experiencing the posterior-collapse phenomenon common in training VAEs for problems like text generation. Experiments show this approach outperforms strong autoregressive and variational baselines on standard text and image datasets.

Learning the latent representation of data in unsupervised fashion is a very interesting process that provides relevant features for enhancing the performance of a classifier. For speech emotion recognition tasks, generating effective features is crucial. Currently, handcrafted features are mostly used for speech emotion recognition, however, features learned automatically using deep learning have shown strong success in many problems, especially in image processing. In particular, deep generative models such as Variational Autoencoders (VAEs) have gained enormous success for generating features for natural images. Inspired by this, we propose VAEs for deriving the latent representation of speech signals and use this representation to classify emotions. To the best of our knowledge, we are the first to propose VAEs for speech emotion classification. Evaluations on the IEMOCAP dataset demonstrate that features learned by VAEs can produce state-of-the-art results for speech emotion classification.

An important problem in the social sciences is estimating the effect of a policy intervention on an outcome over time. When interventions take place at an aggregate level (e.g., city or state), researchers make causal inferences by comparing the post-intervention outcomes for affected units ("treated") against the outcomes of a group of unaffected units ("control"). The synthetic control method (SCM) (Abadie, Diamond, and Hainmueller 2010) has become a popular method for making causal inferences on observational time-series. The method compares a single treated unit outcome with a synthetic control that combines the outcomes of multiple control units on the basis of their pre-intervention similarity with the treated unit. The SCM has several limitations.

Artificial Intelligence agents created by Google's DeepMind has taught itself to walk without receiving any human instructions beyond the task to somehow move forward. This reinforcement learning helps an AI to move efficiently in an unknown terrain without programming.

Machine learning couldn't be hotter, with several heavy hitters offering platforms aimed at seasoned data scientists and newcomers interested in working with neural networks. Among the more popular options is TensorFlow, a machine learning library that Google open-sourced in November 2015. I discussed how the library has become more mature, implemented more algorithms and deployment options, and become easier to program over the preceding year. The best deep learning library had become even better. In this article, I'll give you a very quick gloss on machine learning, introduce you to the basics of TensorFlow, walk you through a few TensorFlow models in the area of image classification, and show you the new high-level APIs.

Word embeddings provide a dense representation of words and their relative meanings. They are an improvement over sparse representations used in simpler bag of word model representations. Word embeddings can be learned from text data and reused among projects. They can also be learned as part of fitting a neural network on text data. In this tutorial, you will discover how to use word embeddings for deep learning in Python with Keras.