Showing papers by "Charles W. Anderson published in 2016"

To Fear or Not to Fear That is the Question: Code Characteristics of a Vulnerable Functionwith an Existing Exploit

Abstract: Not all vulnerabilities are equal. Some recent studies have shown that only a small fraction of vulnerabilities that have been reported has actually been exploited. Since finding and addressing potential vulnerabilities in a program can take considerable time and effort, recently effort has been made to identify code that is more likely to be vulnerable. This paper tries to identify the attributes of the code containing a vulnerability that makes the code more likely to be exploited. We examine 183 vulnerabilities from the National Vulnerability Database for Linux Kernel and Apache HTTP server. These include eighty-two vulnerabilities that have been found to have an exploit according to the Exploit Database. We characterize the vulnerable functions that have no exploit and the ones that have an exploit using eight metrics. The results show that the difference between a vulnerability that has no exploit and the one that has an exploit can potentially be characterized using the chosen software metrics. However, predicting exploitation of vulnerabilities is more complex than predicting just the presence of vulnerabilities and further research is needed using metrics that consider security domain knowledge for enhancing the predictability of vulnerability exploits.

Extending the knowledge of the arabic sentiment classification using a foreign external lexical source

Abstract: This article introduces a methodology for analyzing sentiment in Arabic text using a global foreign lexical source. Our method leverages the available resource in another language such as the SentiWordNet in English to the limited language resource that is Arabic. The knowledge that is taken from the external resource will be injected into the feature model whilethe machine-learning-based classifier is trained. The first step of our method is to build the bag-of-words (BOW) model of the Arabic text. The second step calculates the score of polarity using translation machine technique and English SentiWordNet. The scores for each text will be added to the model in three pairs for objective, positive, and negative. The last step of our method involves training the ML classifier on that model to predict the sentiment of the Arabic text. Our method increases the performance compared with the baseline model that is BOW in most cases. In addition, it seems a viable approach to sentiment analysis in Arabic text where there is limitation of the available resource.

Robust Reinforcement Learning with Relevance Vector Machines

Abstract: Function approximation methods, such as neural networks, radial basis functions, and support vector machines, have been used in reinforcement learning to deal with large state spaces. However, they can become unstable with changes in the samples state distributions and require many samples for good estimations of value functions. Recently, Bayesian approaches to reinforcement learning have shown advantages in the explorationexploitation tradeoff and in lower sampling costs. This paper proposes a novel reinforcement learning framework that uses the relevance vector machines (RVM) as a function approximator, which incrementally accumulates knowledge from experiences based on the sparseness of the RVM model. This gradual knowledge construction process increases the stability and robustness of reinforcement learning by preventing possible forgetting. In addition, RVM’s low sampling costs improve the learning speed. The approach is examined in the popular benchmark problems of pole-balancing and mountain car.

Relevance Vector Sampling for Reinforcement Learning in Continuous Action Space

Abstract: To be applicable to real world problems, much reinforcement learning (RL) research has focused on continuous state spaces with function approximations. Some problems also require continuous actions, but searching for good actions in a continuous action space is problematic. This paper suggests a novel relevance vector sampling approach to action search in an RL framework with relevance vector machines (RVM-RL). We hypothesize that each relevance vector (RV) is placed on the modes of the value approximation surface as the learning converges. From the hypothesis, we select actions in RVs to maximize the estimated state-action values. We report the efficiency of the proposed approach by controlling a simulated octopus arm with RV-sampled actions.

Proceedings of the 1st International Workshop on Robot Learning and Planning (RLP 2016)

Abstract: In this paper we present a simulated annealingbased method for planning efficient paths with a tether which avoid entanglement in an obstacle-filled environment. By evaluating total path cost as a function of both path length and entanglements, a robot can plan a path through multiple points of interest while avoiding becoming entangled in any obstacle. In simulated trials, the robot was able to successfully plan non-entangling paths in an obstacle-filled environment. These results were then validated in pool trials on a SeaBotix vLVB300 underwater vehicle.