A5047105504- Software Testing and Debugging Techniques
- Structural Integrity and Reliability Analysis
- Risk and Safety Analysis
- Software Reliability and Analysis Research
- Multi-Criteria Decision Making
- Distributed systems and fault tolerance
- Formal Methods in Verification
- Software System Performance and Reliability
- Reinforcement Learning in Robotics
- Cloud Computing and Resource Management
- Advanced Data Storage Technologies
- Evolutionary Algorithms and Applications
- Topic Modeling
- Engineering Diagnostics and Reliability
- Radiation Effects in Electronics
- Oil Spill Detection and Mitigation
- Advanced Bandit Algorithms Research
- Adversarial Robustness in Machine Learning
- Recommender Systems and Techniques
- Power Systems and Technologies
- Advanced Text Analysis Techniques
- Data Management and Algorithms
- Human Mobility and Location-Based Analysis
- Corrosion Behavior and Inhibition
- Fault Detection and Control Systems
China Classification Society
2022
Sun Yat-sen University
2020-2022
The Ohio State University
2016-2021
Tianjin University
2018-2021
Collaborative Innovation Centre for Advanced Ship and Deep-Sea Exploration
2019
Guangdong Medical College
2011
China National Nuclear Corporation
1999
Exception mechanism is widely used in cloud systems. This mainly because it separates the error handling code from main business logic. However, huge space of potential conditions and sophisticated logic systems present a big hurdle to correct use exception mechanism. As result, mistakes may lead severe consequences, such as system downtime data loss. To address this issue, communities direly need better understanding exception-related bugs, i.e., eBugs, which are caused by incorrect...
The fuzziness and randomness are significant epistemic uncertainty within the qualitative categorization of two critical factors, frequency severity, which not fully considered in traditional risk matrix. This paper mainly proposes a cloud matrix method for assessment process safety considering expert elicitation. model is employed to provide mathematical expression linguistic variables by its numerical characteristics entropy En hyperentropy He. An adjusted Mamdani inference algorithm...
Network partitions are inevitable in large-scale cloud systems. Despite developer's efforts handling network throughout designing, implementing and testing systems, bugs caused by partitions, i.e., partition bugs, still exist cause severe failures production clusters. It is challenging to expose these because they often require start stop at specific timings.
The Maximum k-plex Problem is an important combinatorial optimization problem with increasingly wide applications. In this paper, we propose a novel strategy, named Dynamic-threshold Configuration Checking (DCC), to reduce the cycling of local search. Due complicated neighborhood relations, all previous search algorithms for spend large amount time in identifying feasible neighbors each step. To further improve performance on dense and challenging instances, Double-attributes Incremental...
Control firmware in unmanned aerial vehicles (UAVs) uses sensors to model and manage flight operations, from takeoff landing flying between waypoints. However, can fail at any time during a flight. If control mishandles sensor failures, UAVs crash, fly away, or suffer other unsafe conditions. In-situ checking finds failures that could lead conditions by systematically failing sensors. the type of failure its timing within affect manifestation, creating large search space. We propose Avis, an...
Hai Wan, Haicheng Chen, Jianfeng Du, Weilin Luo, Rongzhen Ye. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.
Recommender systems are widely used to help users locate information by suggesting goods that best match users' needs and preferences. Most existing recommender built implementing supervised learning-based algorithms. Unfortunately, the large search space with lots of local optimums makes it challenging for apply in more new upcoming recommendation scenarios, such as novel item-list scenario proposed FUXI AI Lab, Netease. It requires recommend an item list each step, where is one thousands...
Control firmware in unmanned aircraft systems (UAS) manage the subsystems for in-flight dynamics, navigation and sensors. Computer on-board on gateway machines can now support rich features control firmware, such as GPS-driven waypoint missions autonomy. However, source code behind harbor software bugs whose symptoms are detectable only during flight. Often, UAS have serious that lead to dangerous situations. We studied previously reported open-source repositories of ArduPilot PX4, two...
Software should behave correctly even in adverse conditions. Particularly, we study the problem of automated validation crash consistency, i.e., file system data safety when systems crash. Existing work requires non-trivial manual efforts specifying checking scripts and workloads, which is an obstacle for software developers. Therefore, propose C3, a novel approach that makes consistency as easy pressing single button. With program input, C3 automatically reports inconsistent sites. not only...
Social relationship understanding infers existing social relationships among individuals in a given scenario, which has been demonstrated to have wide range of practical value reality.However, methods infer the each person pair isolation, without considering context-aware information for pairs same scenario.The exists extensively reality, that is, different simple scenario are always related other.For instance, if most relationship, "friends", then other high probability being "friends" or...
Reinforcement learning competitions advance the field by providing appropriate scope and support to develop solutions toward a specific problem. To promote development of more broadly applicable methods, organizers need enforce use general techniques, sample-efficient reproducibility results. While beneficial for research community, these restrictions come at cost -- increased difficulty. If barrier entry is too high, many potential participants are demoralized. With this in mind, we hosted...
Control firmware in unmanned aerial vehicles (UAVs) uses sensors to model and manage flight operations, from takeoff landing flying between waypoints. However, can fail at any time during a flight. If control mishandles sensor failures, UAVs crash, fly away, or suffer other unsafe conditions. In-situ checking finds failures that could lead conditions by systematically failing sensors. the type of failure its timing within affect manifestation, creating large search space. We propose Avis, an...