Distributed large model training is intensively time and resource consuming. Failures during the long training period are often inevitable, and can incur substantial recovery costs. Checkpointing has been the standard fault tolerance approach, which periodically stores the latest model states at remote persistent storage. This process can be time-consuming due to limited network bandwidth, and adversely affects training throughput. In-memory checkpointing addresses this issue by saving checkpoint data into host memory instead of remote storage. However, host memory is non-persistent, and may not provide sufficient resilience in case of machine failure. We propose ECCheck, a novel in-memory checkpoint system that employs erasure coding to enhance fault tolerance in distributed deep neural network training. ECCheck advocates serialization-free encoding and decoding in model checkpointing. Several techniques are proposed to minimize computation and communication overhead incurred by erasure coding. Extensive experiments demonstrate that ECCheck achieves superior fault tolerance compared to state-of-the-art solutions, while maintaining high checkpointing frequency, low checkpointing stalls, and fast recovery from failures.

Neural networks often tend to rely on bias features that have strong but spurious correlations with the target labels for decision-making, leading to poor performance on data that does not adhere to these correlations. Early debiasing methods typically construct an unbiased optimization objective based on the labels of bias features. Recent work assumes that bias label is unavailable and usually trains two models: a biased model to deliberately learn bias features for exposing data bias, and a target model to eliminate bias captured by the bias model. In this paper, we first reveal that previous biased models fit target labels, which resulted in failing to expose data bias. To tackle this issue, we propose poisoner, which utilizes data poisoning to embed the biases learned by biased models into the poisoned training data, thereby encouraging the models to learn more biases. Specifically, we couple data poisoning and model training to continuously prompt the biased model to learn more bias. By utilizing the biased model, we can identify samples in the data that contradict these biased correlations. Subsequently, we amplify the influence of these samples in the training of the target model to prevent the model from learning such biased correlations. Experiments show the superior debiasing performance of our method.

The prosperity of knowledge graphs, as well as related downstream applications, has raised the urgent need for knowledge graph completion techniques that fully support knowledge graph reasoning tasks, especially under the circumstance of training data scarcity. Although large efforts have been made on solving this challenge via few-shot learning tools, they mainly focus on simply aggregating entity neighbors to represent few-shot references, whereas the enhancement from latent semantic correlation within neighbors has been largely ignored. To that end, in this article, we propose a novel few-shot learning solution named SIM, a S emantic I nteraction M atching network that applies a Transformer framework to enhance the entity representation with capturing semantic interaction between entity neighbors. Specifically, we first design an entity-relation fusion module to adaptively encode neighbors with incorporating relation representation. Along this line, Transformer layers are integrated to capture latent correlation within neighbors, as well as the semantic diversification of the support set. Finally, a similarity score is attentively estimated with the attention mechanism. Extensive experiments on two public benchmark datasets demonstrate that our model outperforms a variety of state-of-the-art methods by a significant margin.

Few-shot Named Entity Recognition (NER) task, which aims to identify and classify entities from different domains with limited training samples, has long been treated as a basic step for knowledge graph (KG) construction. Great efforts have been made on this task with competitive performance, however, they usually treat the two subtasks, namely span detection and type classification, as mutually independent, and the integrity and correlation between subtasks have been largely ignored. Moreover, prior arts may fail to absorb the coarse-grained features of entities, resulting in a semantic-insufficient representation of entity types. To that end, in this paper, we propose a Hierarchical Enhancing ProtoNet (HEProto) based on multi-task learning, which is utilized to jointly learn these two subtasks and model their correlation. Specifically, we adopt contrastive learning to enhance the span boundary information and the type semantic representations in these two subtasks. Then, the hierarchical prototypical network is designed to leverage the coarse-grained information of entities in the type classification stage, which could help the model to better learn the fine-grained semantic representations. Along this line, we construct a similarity margin loss to reduce the similarity between fine-grained entities and other irrelevant coarse-grained prototypes. Finally, extensive experiments on the Few-NERD dataset prove that our solution outperforms competitive baseline methods. The source code of HEProto is available at \hrefhttps://github.com/fanshu6hao/HEProto https://github.com/fanshu6hao/HEProto.

Person search has long been treated as a crucial and challenging task to support deeper insight in personalized summarization and personality discovery. Traditional methods, e.g., person re-identification and face recognition techniques, which profile video characters based on visual information, are often limited by relatively fixed poses or small variation of viewpoints and suffer from more realistic scenes with high motion complexity (e.g., movies). At the same time, long videos such as movies often have logical story lines and are composed of continuously developmental plots. In this situation, different persons usually meet on a specific occasion, in which informative social cues are performed. We notice that these social cues could semantically profile their personality and benefit person search task in two aspects. First, persons with certain relationships usually co-occur in short intervals; in case one of them is easier to be identified, the social relation cues extracted from their co-occurrences could further benefit the identification for the harder ones. Second, social relations could reveal the association between certain scenes and characters (e.g., classmate relationship may only exist among students), which could narrow down candidates into certain persons with a specific relationship. In this way, high-level social relation cues could improve the effectiveness of person search. Along this line, in this article, we propose a social context-aware framework, which fuses visual and social contexts to profile persons in more semantic perspectives and better deal with person search task in complex scenarios. Specifically, we first segment videos into several independent scene units and abstract out social contexts within these scene units. Then, we construct inner-personal links through a graph formulation operation for each scene unit, in which both visual cues and relation cues are considered. Finally, we perform a relation-aware label propagation to identify characters’ occurrences, combining low-level semantic cues (i.e., visual cues) and high-level semantic cues (i.e., relation cues) to further enhance the accuracy. Experiments on real-world datasets validate that our solution outperforms several competitive baselines.

For a long time, PageRank has been widely used for authority computation and has been adopted as a solid baseline for evaluating social influence related applications. However, when measuring the authority of network nodes, the traditional PageRank method does not take the nodes’ prior knowledge into consideration. Also, the connection between PageRank and social influence modeling methods is not clearly established. To that end, this article provides a focused study on understanding PageRank as well as the relationship between PageRank and social influence analysis. Along this line, we first propose a linear social influence model and reveal that this model generalizes the PageRank-based authority computation by introducing some constraints. Then, we show that the authority computation by PageRank can be enhanced if exploiting more reasonable constraints (e.g., from prior knowledge). Next, to deal with the computational challenge of linear model with general constraints, we provide an upper bound for identifying nodes with top authorities. Moreover, we extend the proposed linear model for better measuring the authority of the given node sets, and we also demonstrate the way to quickly identify the top authoritative node sets. Finally, extensive experimental evaluations on four real-world networks validate the effectiveness of the proposed linear model with respect to different constraint settings. The results show that the methods with more reasonable constraints can lead to better ranking and recommendation performance. Meanwhile, the upper bounds formed by PageRank values could be used to quickly locate the nodes and node sets with the highest authorities.

Recent years have witnessed the boom of online sharing media contents, which raise significant challenges in effective management and retrieval. Though a large amount of efforts have been made, precise retrieval on video shots with certain topics has been largely ignored. At the same time, due to the popularity of novel time-sync comments, or so-called "bullet-screen comments", video semantics could be now combined with timestamps to support further research on temporal video labeling. In this paper, we propose a novel video understanding framework to assign temporal labels on highlighted video shots. To be specific, due to the informal expression of bullet-screen comments, we first propose a temporal deep structured semantic model (T-DSSM) to represent comments into semantic vectors by taking advantage of their temporal correlation. Then, video highlights are recognized and labeled via semantic vectors in a supervised way. Extensive experiments on a real-world dataset prove that our framework could effectively label video highlights with a significant margin compared with baselines, which clearly validates the potential of our framework on video understanding, as well as bullet-screen comments interpretation.

In this paper, we present a systematic perspective study on choosing and evaluating the initial seed items that will be recommended to the cold start users. We first construct an item consumption correlation network to capture the existing users' general consumption behaviors. Then, we formalize initial items recommendation as the influential seed set selection problem. Along this line, we present several methods, each of which selects seed items according to different rules. Finally, the experimental results on two real-world data sets verify that with different seed items, the users' consumption numbers will be quite different. Meanwhile, the results also provide many deep insights into these selection methods and their recommended seed items.