Deep Neural Networks (DNNs) have recently been achieving state-of-the-art performance on a variety of computer vision related tasks. However, their computational cost limits their ability to be implemented in embedded systems with restricted resources or strict latency constraints. Model compression has therefore been an active field of research to overcome this issue. On the other hand, DNNs typically require massive amounts of labeled data to be trained. This represents a second limitation to their deployment. Domain Adaptation (DA) addresses this issue by allowing to transfer knowledge learned on one labeled source distribution to a target distribution, possibly unlabeled. In this paper, we investigate on possible improvements of compression methods in DA setting. We focus on a compression method that was previously developed in the context of a single data distribution and show that, with a careful choice of data to use during compression and additional regularization terms directly related to DA objectives, it is possible to improve compression results. We also show that our method outperforms an existing compression method studied in the DA setting by a large margin for high compression rates. Although our work is based on one specific compression method, we also outline some general guidelines for improving compression in DA setting.

However, the existing DL models are highly vulnerable to adversarial examples [55, 20], which are maliciously generated by an adversary to make a model produce erroneous predictions. As DL models have been integrated into various security-sensitive applications ( e.g., autonomous driving, healthcare, and finance), the study of the adversarial robustness issue has attracted increasing attention with an enormous number of adversarial attack and defense methods proposed. Therefore, it is crucial to conduct correct and rigorous evaluations of these methods for understanding their pros and cons, comparing their performance, and providing insights for building new methods [6]. The research on adversarial robustness is faced with an "arms race " between attacks and defenses, i.e ., a defense method proposed to prevent the existing attacks was soon evaded by new attacks, and vice versa [7, 8, 23, 1, 57, 67]. For instance, defensive distillation [43] was proposed to improve adversarial robustness, but was later shown to be ineffective against a strong attack [8].

Deep neural networks and in particular, deep neural classifiers have become an integral part of many modern applications. Despite their practical success, we still have limited knowledge of how they work and the demand for such an understanding is evergrowing. In this regard, one crucial aspect of deep neural network classifiers that can help us deepen our knowledge about their decision-making behavior is to investigate their decision boundaries. Nevertheless, this is contingent upon having access to samples populating the areas near the decision boundary. To achieve this, we propose a novel approach we call Deep Decision boundary Instance Generation (DeepDIG). DeepDIG utilizes a method based on adversarial example generation as an effective way of generating samples near the decision boundary of any deep neural network model. Then, we introduce a set of important principled characteristics that take advantage of the generated instances near the decision boundary to provide multifaceted understandings of deep neural networks. We have performed extensive experiments on multiple representative datasets across various deep neural network models and characterized their decision boundaries.

During training, the weights of a Deep Neural Network (DNN) are optimized from a random initialization towards a nearly optimum value minimizing a loss function. Only this final state of the weights is typically kept for testing, while the wealth of information on the geometry of the weight space, accumulated over the descent towards the minimum is discarded. In this work we propose to make use of this knowledge and leverage it for computing the distributions of the weights of the DNN. This can be further used for estimating the epistemic uncertainty of the DNN by sampling an ensemble of networks from these distributions. T o this end we introduce a method for tracking the trajectory of the weights during optimization, that does not require any changes in the architecture nor on the training procedure. W e evaluate our method on standard classification and regression benchmarks, and on out-of-distribution detection for classification and semantic segmentation. W e achieve competitive results, while preserving computational efficiency in comparison to other popular approaches.

Binarization is an attractive strategy for implementing lightweight Deep Convolutional Neural Networks (CNNs). Despite the unquestionable savings offered, memory footprint above all, it may induce an excessive accuracy loss that prevents a widespread use. This work elaborates on this aspect introducing TentacleNet, a new template designed to improve the predictive performance of binarized CNNs via parallelization. Inspired by the ensemble learning theory, it consists of a compact topology that is end-to-end trainable and organized to minimize memory utilization. Experimental results collected over three realistic benchmarks show TentacleNet fills the gap left by classical binary models, ensuring substantial memory savings w.r.t. state-of-the-art binary ensemble methods.

Object perception is a fundamental sub-field of Computer Vision, covering a multitude of individual areas and having contributed high-impact results. While Machine Learning has been traditionally applied to address related problems, recent works also seek ways to integrate knowledge engineering in order to expand the level of intelligence of the visual interpretation of objects, their properties and their relations with their environment. In this paper, we attempt a systematic investigation of how knowledge-based methods contribute to diverse object perception tasks. We review the latest achievements and identify prominent research directions.

We use synthetic data and a reinforcement learning algorithm to train a system controlling a full-size real-world vehicle in a number of restricted driving scenarios. The driving policy uses RGB images as input. We analyze how design decisions about perception, control and training impact the real-world performance.

Recently OpenAI announced the launch of its 1.5 billion parameter language model GPT-2. GPT-2 has been in the news as the scary AI text generator with potential threats regarding fake news stories, and so on. But now we have'GPTrue or False' browser extension that displays the GPT-2 Log Probability of selected portions of text. This browser extension allows you to select text on a website and finds out what you selected is written using OpenAI's GPT-2 A.I. model. GPTrue or False is available both for Chrome and Firefox.

We write a lot about open problems in Natural Language Processing. We complain a lot when working on NLP projects. We pick on inaccuracies and blatant errors of different models. But what we need to admit is that NLP has already changed and new models have solved the problems that may still linger in our memory. One of such drastic developments is the launch of Google's Bidirectional Encoder Representations from Transformers, or BERT model -- the model that is called the best NLP model ever based on its superior performance over a wide variety of tasks.

With the presence of artificial intelligence (AI) everywhere and the increased use of deep learning (DL), many security practitioners are being hooked into believing that these approaches are the solution for the security challenges. Nevertheless, like any tool, AI is a double-edged sword that can be used as a security solution or as a weapon by hackers. In fact, many security researchers and industry have told AI will be the biggest ally of security. Moreover, we can see that by the increased number of companies that merging AI and Cybersecurity to keep us safe. But has anyone ever thought that these same techniques can be applied to improve the tools and methods used by hackers?