We build on abduction-based explanations for machine learning and develop a method for computing local explanations for neural network models in natural language processing (NLP). Our explanations comprise a subset of the words of the input text that satisfies two key features: optimality w.r.t. a user-defined cost function, such as the length of explanation, and robustness, in that they ensure prediction invariance for any bounded perturbation in the embedding space of the left-out words. We present two solution algorithms, respectively based on implicit hitting sets and maximum universal subsets, introducing a number of algorithmic improvements to speed up convergence of hard instances. We show how our method can be configured with different perturbation sets in the embedded space and used to detect bias in predictions by enforcing include/exclude constraints on biased terms, as well as to enhance existing heuristic-based NLP explanation frameworks such as Anchors. We evaluate our framework on three widely used sentiment analysis tasks and texts of up to 100 words from SST, Twitter and IMDB datasets, demonstrating the effectiveness of the derived explanations.

No abstract

No abstract

No abstract

For many years various types of devices equipped with sensors have guaranteed proper work in a huge amount of machines and systems. For the proper operation of sensors, devices, and complex systems, we need secure communication. Security protocols (SP) in this case, guarantee the achievement of security goals. However, the design of SP is not an easy process. Sometimes SP cannot realise their security goals because of errors in their constructions and need to be investigated and verified in the case of their correctness. Now SP uses often time primitives due to the necessity of security dependence on the passing of time. In this work, we propose and investigate the SAT-and SMT-based formal verification methods of SP used in communication between devices equipped with sensors. For this, we use a formal model based on networks of communicating timed automata. Using this, we show how the security property of SP dedicated to the sensors world can be verified. In our work, we investigate such timed properties as delays in the network and lifetimes. The delay in the network is the lower time constraint related to sending the message. Lifetime is an upper constraint related to the validity of the timestamps generated for the transmitted messages.

This article provides a comprehensive and up-to-date overview of the repositories that contain color fundus images. We analyzed them regarding availability and legality, presented the datasets’ characteristics, and identified labeled and unlabeled image sets. This study aimed to complete all publicly available color fundus image datasets to create a central catalog of available color fundus image datasets.

No abstract

No abstract