InterpNET
Humans are able to explain their reasoning. On the contrary, deep neural networks are not. This paper attempts to bridge this gap by introducing a new way to design interpretable neural networks for classification, inspired by physiological evidence of the human visual system’s inner-workings. This paper proposes a neural network design paradigm, termed InterpNET, which can be combined with any existing classification architecture to generate natural language explanations of the classifications. The success of the module relies on the assumption that the network’s computation and reasoning is represented in its internal layer activations. While in principle InterpNET could be applied to any existing classification architecture, it is evaluated via an image classification and explanation task. Experiments on a CUB bird classification and explanation dataset show qualitatively and quantitatively that the model is able to generate high-quality explanations. While the current state-of-the-art METEOR score on this dataset is 29.2, InterpNET achieves a much higher METEOR score of 37.9. …

Neural Machine Translation (NMT)
Neural machine translation (NMT) is the approach to machine translation in which a large neural network is trained to maximize translation performance. It is a radical departure from the phrase-based statistical translation approaches, in which a translation system consists of subcomponents that are separately optimized. The artificial neural network (ANN) is a model inspired by the functional aspects and structure of the brain’s biological neural networks. With use of ANN, it is possible to execute a number of tasks, such as classification, clustering, and prediction, using machine learning techniques like supervised or reinforced learning to learn or adjust net connections. A bidirectional recurrent neural network (RNN), known as an encoder, is used by the neural network to encode a source sentence for a second RNN, known as a decoder, that is used to predict words in the target language. NMT models are inspired by deep representation learning. They require only a fraction of the memory needed by traditional statistical machine translation (SMT) models. Furthermore, unlike conventional translation systems, each and every component of the neural translation model is trained jointly to maximize the translation performance. When a new neural network is created, it is trained for certain domains or applications. Once an automatic learning mechanism is established, the network practices. With time it starts operating according to its own judgment, turning into an ‘expert’. …

EagleEye
Deep neural networks (DNNs) are inherently vulnerable to adversarial inputs: such maliciously crafted samples trigger DNNs to misbehave, leading to detrimental consequences for DNN-powered systems. The fundamental challenges of mitigating adversarial inputs stem from their adaptive and variable nature. Existing solutions attempt to improve DNN resilience against specific attacks; yet, such static defenses can often be circumvented by adaptively engineered inputs or by new attack variants. Here, we present EagleEye, an attack-agnostic adversarial tampering analysis engine for DNN-powered systems. Our design exploits the {\em minimality principle} underlying many attacks: to maximize the attack’s evasiveness, the adversary often seeks the minimum possible distortion to convert genuine inputs to adversarial ones. We show that this practice entails the distinct distributional properties of adversarial inputs in the input space. By leveraging such properties in a principled manner, EagleEye effectively discriminates adversarial inputs and even uncovers their correct classification outputs. Through extensive empirical evaluation using a range of benchmark datasets and DNN models, we validate EagleEye’s efficacy. We further investigate the adversary’s possible countermeasures, which implies a difficult dilemma for her: to evade EagleEye’s detection, excessive distortion is necessary, thereby significantly reducing the attack’s evasiveness regarding other detection mechanisms. …