Coarse-grained reconfigurable architectures (CGRAs) have drawn increasing attention due to their flexibility and energy efficiency. Data flow graphs (DFGs) are often mapped onto CGRAs for acceleration. The problem of DFG mapping is challenging due to the diverse structures from DFGs and constrained hardware from CGRAs. Consequently, it is difficult to find a valid and high quality solution simultaneously. Inspired from the great progress in deep reinforcement learning for AI problems, we consider building methods that learn to map DFGs onto spatially-programmed CGRAs directly from experiences. We propose RLMap, a solution that formulates DFG mapping on CGRA as an agent in reinforcement learning (RL), which unifies placement, routing and PE insertion by interchange actions of the agent. Experimental results show that RLMap performs comparably to state-of-the-art heuristics in mapping quality, adapts to different architecture and converges quickly.

Flight safety is a key issue in the aviation industry. Recently, with the prevalence of flight data recording systems, some deep learning-based studies have been devoted to predicting safety incidents based on flight data. However, these studies, although they exhibit higher prediction accuracy, have largely neglected the interpretability analysis of safety incidents which is of great concern to airlines and pilots. To address this issue, we define flight safety prediction as a multiscale time series classification problem and propose an interpretable model named IMTCN to provide both accurate predictions and high interpretability of flight safety. First, multiple temporal convolutional networks (TCNs) are utilized to capture local representations and long effective histories from multivariate flight data. Because different flight parameters are collected with diverse sampling frequencies, multiple TCNs are used to handle these parameters separately. Then, we creatively adapt the class activation mapping (CAM) method, which has been used for interpretation in image classification, and combine it with the TCN to provide flight data interpretability. The established model can pinpoint key flight parameters and corresponding moments that contribute most to safety incidents. Experimental results on a real-world dataset with 37,943 Airbus A320 aircraft flights show that our model outperforms the baselines on the task of exceedance classification and prediction 2 seconds and 4 seconds in advance, and case studies demonstrate its superb interpretability for flight safety analysis.

Influence maximization problem has been extensively studied in recent years. It aims at finding a seed set consisting of $k$ vertices from a network, so that their collective influence spread can be maximized. However, existing works have neglected the user admission behavior where the successfully influenced individuals do not always disseminate the messages to their neighbors, and we show that ignoring admission behavior may incur significant bias to existing methods. Based on this observation, a new admitted influence maximization (AIM) problem is proposed by taking the admission behavior into consideration. Specifically, we first propose two new diffusion models, i.e., AIC and ALT, by extending traditional independent cascade and linear threshold models. Then, a novel edge coloring scheme is proposed to theoretically prove the NP-hardness of the AIM problem and the submodularity of its objective function. Based on the theoretical findings, a greedy algorithm is given to find a $(1-1/e-\epsilon )$ approximate solution. To further improve the algorithm efficiency and handle large-scale networks, an efficient algorithm ARIS based on valid path reverse influence sampling is proposed, which could also ensure a $(1-1/e-\epsilon )$ approximate solution. Experimental results on several large-scale real networks exhibit the effectiveness and efficiency of our proposed algorithm.