Security vulnerabilities play a vital role in network security system. Fuzzing technology is widely used as a vulnerability discovery technology to reduce damage in advance. However, traditional fuzz testing faces many challenges, such as how to mutate input seed files, how to increase code coverage, and how to bypass the format verification effectively. Therefore machine learning techniques have been introduced as a new method into fuzz testing to alleviate these challenges. This paper reviews the research progress of using machine learning techniques for fuzz testing in recent years, analyzes how machine learning improves the fuzzing process and results, and sheds light on future work in fuzzing. Firstly, this paper discusses the reasons why machine learning techniques can be used for fuzzing scenarios and identifies five different stages in which machine learning has been used. Then this paper systematically studies machine learning-based fuzzing models from five dimensions of selection of machine learning algorithms, pre-processing methods, datasets, evaluation metrics, and hyperparameters setting. Secondly, this paper assesses the performance of the machine learning techniques in existing research for fuzz testing. The results of the evaluation prove that machine learning techniques have an acceptable capability of prediction for fuzzing. Finally, the capability of discovering vulnerabilities both traditional fuzzers and machine learning-based fuzzers is analyzed. The results depict that the introduction of machine learning techniques can improve the performance of fuzzing. We hope to provide researchers with a systematic and more in-depth understanding of fuzzing based on machine learning techniques and provide some references for this field through analysis and summarization of multiple dimensions.
Abstract. We present the results of pollen analyses from a 1105-cm-long sediment core from Wuxu Lake in southwestern China, which depict the variations of the East Asian winter monsoon (EAWM) and the Indian summer monsoon (ISM) during the last 12.3 ka. During the period of 12.3 to 11.3 cal ka BP, the dominance of Betula forest and open alpine shrub and meadow around Wuxu Lake indicates a climate with relatively cold winters and dry summers, corresponding to the Younger Dryas event. Between 11.3 and 10.4 cal ka BP, further expansion of Betula forest and the retreat of alpine shrubs and meadows reflect a greater seasonality with cold winters and gradually increasing summer precipitation. From 10.4 to 4.9 cal ka BP, the dense forest understory, together with the gradual decrease in Betula forest and increase in Tsuga forest, suggest that the winters became warmer and summer precipitation was at a maximum, corresponding to the Holocene climatic optimum. Between 4.9 and 2.6 cal ka BP, Tsuga forest and alpine shrubs and meadows expanded significantly, reflecting relatively warm winters and decreased summer precipitation. Since 2.6 cal ka BP, reforestation around Wuxu Lake indicates a renewed strengthening of the ISM in the late Holocene; however, the vegetation in the catchment may also have been affected by grazing activity during this period. The results of our study are generally consistent with previous findings; however, the timing and duration of the Holocene climatic optimum from different records are inconsistent, reflecting real contrast in local rainfall response to the ISM. Overall, the EAWM is broadly in-phase with the ISM on the orbital timescale, and both monsoons exhibit a trend of decreasing strength from the early to late Holocene, reflecting the interplay of solar insolation receipt between the winter and summer seasons and El Niño Southern Oscillation strength in the tropical Pacific.
peer-reviewed The study of gas flows in microchannels has received considerably more attention in the literature from a simulation perspective than an experimental. The majority of the experimental work has emphasis on the global measurements at the inlet or exit of the microchannel instead locally along it. In this paper some efforts were made to measure the pressure drop along T-shaped micro channel by using interferometry. The two side channels were served as gas entrances and they were both open to air and the channel outlet was being vacuumed during experiments. A Mach-Zehnder interference microscopy was built for the measurement of gas pressure drop along the mixing channel. Some points along the mixing channel were selected for interferometric measurements. Simulations were first developed in unsteady condition by using Ansys Fluent to verify the nonexistence of transient phenomena of gas flow in the defined condition and then run again in steady condition to get the theoretical pressure drop that was would be used for comparison with experimental results. PUBLISHED peer-reviewed
It is hard to select and read suitable documents due to the rapidly growing number of scholarly documents. Keyphrases can be considered as the gist of a document so that a researcher can select the documents that they want using keyphrase queries. However, there are also many scholarly documents without any keyphrases tagged by the authors or other researchers. Automatic keyphrase extraction can help researchers to quickly extract keyphrases. This paper proposed an unsupervised approach for keyphrase extraction using graph-based ranking and topic-based clustering under the assumption that we only use the within-collection resources. We use graph-based ranking to describe the relevance between two words and topic-based clustering to embed semantical information into words. In this paper, we assume that each word has its own meaning, and each meaning can be considered as a topic, though we know nothing about these meanings. We use topic-based clustering to assign the “correct meaning” to the “correct word”. In addition, by taking the relevance among phrases into consideration and only using within-collection resources, we can use the graph-based ranking in our approach. The edges in a graph that are built for phrases can describe the hidden relevance between two phrases, and the weights that are set for edges can measure the connection between two phrases. Then, after using the position feature, our approach consists of an enhanced graph-based ranking and a topic-based clustering. The experiments are run on four datasets: KDD, WWW, GSN and ACM. The results indicate that our approach has better performance than the state-of-the-art methods.
This paper describes our system submitted to task 4 of SemEval 2020: Commonsense Validation and Explanation (ComVE) which consists of three sub-tasks. The task is to directly validate the given sentence whether or not to make sense and require the model to explain it. Based on BERT architecture with the multi-task setting, we propose an effective and interpretable “Explain, Reason and Predict” (ERP) system to solve the three sub-tasks about commonsense: (a) Validation, (b) Reasoning, and (c) Explanation. Inspired by cognitive studies of common sense, our system first generates a reason or understanding of the sentences and then choose which one statement makes sense, which is achieved by multi-task learning. During the post-evaluation, our system has reached 92.9% accuracy in subtask A (rank 11), 89.7% accuracy in subtask B (rank 9), and BLEU score of 12.9 in subtask C (rank 8).
Late March is the end of the cicada season in New Zealand. But the summer has been unusually wet in Auckland, and the insects' constant singing is still going strong in Henderson Park, where a larger-than-usual crowd of visitors has gathered for the 2017 edition of Bioblitz. The goal of this event