Despite the utility of Maximum Satisfiability (MaxSAT) in discrete optimization, developing iterative workflows remains cumbersome due to fragmented, low-level solver APIs. We present Hermax, a unified Python library and modelling compiler for MaxSAT. Hermax provides an IPAMIR interface that exposes incremental solving, assumptions, and weight updates through a single API across many incremental and non-incremental backends. Furthermore, it introduces a compiler with Constraint Programming primitives that translates high-level models directly into optimized CNF/WCNF through eager evaluation. This compiler allows automatic optimizations like integer ladder graph encoding that bypasses Pseudo-Boolean formulation when possible. Together, these features enable rapid prototyping and production grade optimization directly from Python across major platforms and hardware architectures.

Machine learning has become an appealing signature-less approach to detect and classify malware because of its ability to generalize to never-before-seen samples and to handle large volumes of data. While traditional feature-based approaches rely on the manual design of hand-crafted features based on experts’ knowledge of the domain, deep learning approaches replace the manual feature engineering process by an underlying system, typically consisting of a neural network with multiple layers, that perform both feature learning and classification altogether. However, the combination of both approaches could substantially enhance detection systems. In this paper we present an hybrid approach to address the task of malware classification by fusing multiple types of features defined by experts and features learned through deep learning from raw data. In particular, our approach relies on deep learning to extract N-gram like features from the assembly language instructions and the bytes of malware, and texture patterns and shapelet-based features from malware’s grayscale image representation and structural entropy, respectively. These deep features are later passed as input to a gradient boosting model that combines the deep features and the hand-crafted features using an early-fusion mechanism. The suitability of our approach has been evaluated on the Microsoft Malware Classification Challenge benchmark and results show that the proposed solution achieves state-of-the-art performance and outperforms gradient boosting and deep learning methods in the literature.

Malware detection and classification is a challenging problem and an active area of research. Traditional machine learning methods depend almost entirely on the ability to extract a set of discriminative features into which characterize malware. However, this feature engineering process is very time consuming. On the contrary, deep learning methods replace manual feature engineering by a system that performs both feature extraction and classification from raw data at once. Despite that, a major shortfall of these methods is their inhability to consider multiple disparate sources of information when performing classification, leading them to perform poorly when compared to multimodal approaches. In this work, we introduce Orthrus, a new bimodal approach to categorize malware into families based on deep learning. Orthrus combines two modalities of data: (1) the byte sequence representing the malware's binary content, and (2) the assembly language instructions extracted from the assembly language source code of malware, and performs automatic feature learning and classification with a convolutional neural network. The idea is to benefit from multiple feature types to reflect malware's characteristics. The experiments carried on the Microsoft Malware Classification Challenge dataset show that our proposed solution achieves higher classification performance than deep learning approaches in the literature and n-gram based methods.

The struggle between security analysts and malware developers is a never-ending battle with the complexity of malware changing as quickly as innovation grows. Current state-of-the-art research focus on the development and application of machine learning techniques for malware detection due to its ability to keep pace with malware evolution. This survey aims at providing a systematic and detailed overview of machine learning techniques for malware detection and in particular, deep learning techniques. The main contributions of the paper are: (1) it provides a complete description of the methods and features in a traditional machine learning workflow for malware detection and classification, (2) it explores the challenges and limitations of traditional machine learning and (3) it analyzes recent trends and developments in the field with special emphasis on deep learning approaches. Furthermore, (4) it presents the research issues and unsolved challenges of the state-of-the-art techniques and (5) it discusses the new directions of research. The survey helps researchers to have an understanding of the malware detection field and of the new developments and directions of research explored by the scientific community to tackle the problem.

Malware detection and classification is a challenging problem and an active area of research. Particular challenges include how to best treat and preprocess malicious executables in order to feed machine learning algorithms. Novel approaches in the literature treat an executable as a sequence of bytes or as a sequence of assembly language instructions. However, in those approaches the hierarchical structure of programs is not taken into consideration. An executable exhibits various levels of spatial correlation. Adjacent code instructions are correlated spatially but that is not necessarily the case. Function calls and jump commands transfer the control of the program to a different point in the instruction stream. Furthermore, these discontinuities are maintained when treating the binary as a sequence of byte values. In addition, functions might be arranged randomly if addresses are correctly reorganized. To address these issues we propose a Hierarchical Convolutional Network (HCN) for malware classification. It has two levels of convolutional blocks applied at the mnemonic-level and at the function-level, enabling us to extract n-gram like features from both levels when constructing the malware representation. We validate our HCN method on the dataset released for the Microsoft Malware Classification Challenge, outperforming almost every deep learning method in the literature.

The number of malicious programs has grown both in number and in sophistication. Analyzing the malicious intent of vast amounts of data requires huge resources and thus, effective categorization of malware is required. In this paper, the content of a malicious program is represented as an entropy stream, where each value describes the amount of entropy of a small chunk of code in a specific location of the file. Wavelet transforms are then applied to this entropy signal to describe the variation in the entropic energy. Motivated by the visual similarity between streams of entropy of malicious software belonging to the same family, we propose a file agnostic deep learning approach for categorization of malware. Our method exploits the fact that most variants are generated by using common obfuscation techniques and that compression and encryption algorithms retain some properties present in the original code. This allows us to find discriminative patterns that almost all variants in a family share. Our method has been evaluated using the data provided by Microsoft for the BigData Innovators Gathering Anti-Malware Prediction Challenge, and achieved promising results in comparison with the State of the Art.

One of the most challenging problems on modern urban planning and one of the goals to be solved for smart city design is that of urban waste disposal. Given urban population growth, and that the amount of waste generated by each of us citizens is also growing, the total amount of waste to be collected and treated is growing dramatically (EPA 2011), becoming one sensitive issue for local governments. A modern technique for waste collection that is steadily being adopted is automated vacuum waste collection. This technology uses air suction on a closed network of underground pipes to move waste from the collection points to the processing station, reducing greenhouse gas emissions as well as inconveniences to citizens (odors, noise, . . . ) and allowing better waste reuse and recycling. This technique is open to optimize energy consumption because moving huge amounts of waste by air impulsion requires a lot of electric power. The described problem challenge here is, precisely, that of organizing and scheduling waste collection to minimize the amount of energy per ton of collected waste in such a system via the use of Artificial Intelligence techniques. This kind of problems are an inviting opportunity to showcase the possibilities that AI for Computational Sustainability offers.

Automated vacuum waste collection (AVWC) uses air suction on a closed network of underground pipes to transport waste from the drop off points scattered throughout the city to a central collection point, reducing greenhouse gas emissions and the inconveniences of conventional methods (odors, noise). Since a significant part of the cost of operating AVWC systems is energy consumption, we have started a project, together with a company that builds and installs such systems, with the aim of applying constraint programming technology to schedule the daily emptying sequences of the drop off points in such a way that energy consumption is minimized. In this paper we describe how the problem of deciding the drop off points that should be emptied at a given time can be modeled as a constraint integer programming (CIP) problem. Moreover, we report on experiments using real data from AVWC systems installed in different cities that provide empirical evidence that CIP offers a suitable technology for reducing energy consumption in AVWC.

This paper introduces new solving strategies for the resolution of Pseudo-Boolean Modularity (PBMod) constraints. In particular, we deal with modular arithmetic constraints on Boolean variables. On the one hand, we analyze translations to Pseudo-Boolean (PB) constraints and apply PB solvers. We also look at those PB solvers that have shown that a transformation to the SAT problem can be an effective solving strategy for PB problems. Among the existing translation techniques we focus on the encoding based on a network of sorters. We extend this encoding technique to generate directly a SAT formula from the PBMod constraints. We compare our approach to other standard techniques such as Satisfiability Modulo Theories (SMT) solvers with support for the Quantifier Free Linear Integer Arithmetic (QF_LIA) theory and the GLPK package for Mixed Integer Programming. In order to conduct our experimental investigation we present a generator of random PBMod constraints and study the impact of the several parameters on the hardness of the instances.