Graph similarity measurement, which computes the distance/similarity between two graphs, arises in various graph-related tasks. Recent learning-based methods lack interpretability, as they directly transform interaction information between two graphs into one hidden vector and then map it to similarity. To cope with this problem, this study proposes a more interpretable end-to-end paradigm for graph similarity learning, named Similarity Computation via Maximum Common Subgraph Inference (INFMCS). Our critical insight into INFMCS is the strong correlation between similarity score and Maximum Common Subgraph (MCS). We implicitly infer MCS to obtain the normalized MCS size, with the supervision information being only the similarity score during training. To capture more global information, we also stack some vanilla transformer encoder layers with graph convolution layers and propose a novel permutation-invariant node Positional Encoding. The entire model is quite simple yet effective. Comprehensive experiments demonstrate that INFMCS consistently outperforms state-of-the-art baselines for graph-graph classification and regression tasks. Ablation experiments verify the effectiveness of the proposed computation paradigm and other components. Also, visualization and statistics of results reveal the interpretability of INFMCS.

The reaction center consists of atoms in the product whose local properties are not identical to the corresponding atoms in the reactants. Prior studies on reaction center identification are mainly on semi-templated retrosynthesis methods. Moreover, they are limited to single reaction center identification. However, many reaction centers are comprised of multiple bonds or atoms in reality. We refer to it as the multiple reaction center. This paper presents RCsearcher, a unified framework for single and multiple reaction center identification that combines the advantages of the graph neural network and deep reinforcement learning. The critical insight in this framework is that the single or multiple reaction center must be a node-induced subgraph of the molecular product graph. At each step, it considers choosing one node in the molecular product graph and adding it to the explored node-induced subgraph as an action. Comprehensive experiments demonstrate that RCsearcher consistently outperforms other baselines and can extrapolate the reaction center patterns that have not appeared in the training set. Ablation experiments verify the effectiveness of individual components, including the beam search and one-hop constraint of action space.

Time series analysis is an important and challenging problem in data mining, where time series is a class of temporal data objects. In the classification task, the label is dependent on the features from the last moments. Due to the time dependency, the recurrent neural networks, as one of the prevalent learning-based architectures, take advantage of the relation among history data. The Long Short-Term Memory Network (LSTM) and Gated Recurrent Unit (GRU) are two popular artificial recurrent neural networks used in the field of deep learning. LSTM designed a gate-like method to control the short and long historical information, and GRU simplified those gates to obtain more efficient training. In our work, we propose a new model called as Long Memory Gated Recurrent Unit (LMGRU) based on such two remarkable models, where the reset gate is introduced to reset the stored value of the cell in Long Short-Term Memory (LSTM) model but the forget gate and the input gate are omitted. The experimental results on several time series benchmarks show that LMGRU achieves better effectiveness and efficiency than LSTM and GRU.