Extrapolation over temporal knowledge graph via hyperbolic embedding

Abstract

Predicting potential facts in the future, Temporal Knowledge Graph (TKG) extrapolation remains challenging because of the deep dependence between the temporal association and semantic patterns of facts. Intuitively, facts (events) that happened at different timestamps have different influences on future events, which can be attributed to a hierarchy among not only facts but also relevant entities. Therefore, it is crucial to pay more attention to important entities and events when forecasting the future. However, most existing methods focus on reasoning over temporally evolving facts or mining evolutional patterns from known facts, which may be affected by the diversity and variability of the evolution, and they might fail to attach importance to facts that matter. Hyperbolic geometry was proved to be effective in capturing hierarchical patterns among data, which is considered to be a solution for modelling hierarchical relations among facts. To this end, we propose ReTIN, a novel model integrating real-time influence of historical facts for TKG reasoning based on hyperbolic geometry, which provides low-dimensional embeddings to capture latent hierarchical structures and other rich semantic patterns of the existing TKG. Considering both real-time and global features of TKG boosts the adaptation of ReTIN to the ever-changing dynamics and inherent constraints. Extensive experiments on benchmarks demonstrate the superiority of ReTIN over various baselines. The ablation study further supports the value of exploiting temporal information.

Generation and Characteristics of Construction Noise in Rail Transit Engineering Enclosure Structures

“…The most common symptoms include sleep disturbances [48,49], poor concentration [48], headache, head pressure [48], mental fatigue [50], tiredness, irritability [49], and reduced auditory sensitivity [51]. With the rapid development in the fields of deep learning and artificial intelligence, numerous efficient algorithms have demonstrated outstanding performance in noise monitoring and prediction [52,53], emission sound pressure and sound wave control [54,55], as well as the optimization of noise control systems [56,57]. These novel active noise control methods integrated with artificial intelligence effectively overcome the limitations of traditional passive noise control methods in attenuating low-frequency noise, offering new directions for future construction noise control [58,59].…”

RiQ-KGC: Relation Instantiation Enhanced Quaternionic Attention for Complex-Relation Knowledge Graph Completion

“…RoTH [34] combines hyperbolic reflections and rotations with attention to simultaneously capture hierarchical and logical patterns in a hyperbolic space. ReTIN [35] introduces real-time influence information on historical facts in hyperbolic geometric space to capture potential hierarchical structures and other rich semantic patterns.…”

Rethinking multi‐spatial information for transferable adversarial attacks on speaker recognition systems

“…In the grey-box scenario, attackers optimise the adversarial audio by constantly querying the victim system and obtaining the corresponding score vector [13], which requires frequent access to the victim system and may expose the attacking intent. However, the internal information of the model is usually unknowable in practical scenarios [14][15][16], and the SRS may only output a speaker identity label rather than a score vector. Therefore, the black-box scenario has received more attention from the academic community.…”