Computing Research Repository

Mücke, Jonas; Nawrocki, Marcin; Hiesgen, Raphael; Sattler, Patrick; Zirngibl, Johannes; Carle, Georg; Luxemburk, Jan; Schmidt, Thomas C.; Wählisch, Matthias

doi: 10.1145/3768988pmid: N/A

Abstract:QUIC experiences a rapid adoption since its standardization in 2021, and hypergiants configure their infrastructure to optimize for QUIC performance. In this paper, we introduce a passive measurement method to study both the progressive rollout and individual hypergiant configurations during the last five years. By analyzing backscatter traffic of the UCSD network telescope, we are able to make the following observations. First, Meta, Google, and Cloudflare configure significantly different maximal retransmission numbers and timeouts. Second, we can identify different off-net deployments of hypergiants, using packet features, such as QUIC connection IDs, packet coalescence, and packet lengths. Third, we observe changing hypergiant deployment configurations during our different measurement periods. Fourth, connection IDs can allow further insights into load balancer deployments, such as the number of servers. We bolster our results using two orthogonal measurements: passive recording of QUIC flows and active probing.

Wang, Lianyu; Wang, Meng; Zhang, Daoqiang; Fu, Huazhu

doi: 10.48550/arxiv.2209.01870pmid: N/A

Abstract:Unsupervised domain adaptation (UDA) has attracted considerable attention, which transfers knowledge from a label-rich source domain to a related but unlabeled target domain. Reducing inter-domain differences has always been a crucial factor to improve performance in UDA, especially for tasks where there is a large gap between source and target domains. To this end, we propose a novel style-aware feature fusion method (SAFF) to bridge the large domain gap and transfer knowledge while alleviating the loss of class-discriminative information. Inspired by the human transitive inference and learning ability, a novel style-aware self-intermediate domain (SSID) is investigated to link two seemingly unrelated concepts through a series of intermediate auxiliary synthesized concepts. Specifically, we propose a novel learning strategy of SSID, which selects samples from both source and target domains as anchors, and then randomly fuses the object and style features of these anchors to generate labeled and style-rich intermediate auxiliary features for knowledge transfer. Moreover, we design an external memory bank to store and update specified labeled features to obtain stable class features and class-wise style features. Based on the proposed memory bank, the intra- and inter-domain loss functions are designed to improve the class recognition ability and feature compatibility, respectively. Meanwhile, we simulate the rich latent feature space of SSID by infinite sampling and the convergence of the loss function by mathematical theory. Finally, we conduct comprehensive experiments on commonly used domain adaptive benchmarks to evaluate the proposed SAFF, and the experimental results show that the proposed SAFF can be easily combined with different backbone networks and obtain better performance as a plug-in-plug-out module.

Zhang, Xianqi; Wang, Xingtao; Liu, Xu; Wang, Wenrui; Fan, Xiaopeng; Zhao, Debin

doi: 10.48550/arxiv.2209.04100pmid: N/A

Abstract:Reinforcement Learning (RL) and Imitation Learning (IL) have made great progress in robotic decision-making in recent years. However, these methods show obvious deterioration for new tasks that need to be completed through new combinations of actions. RL methods suffer from reward functions and distribution shifts, while IL methods are limited by expert demonstrations which do not cover new tasks. In contrast, humans can easily complete these tasks with the fragmented knowledge learned from task-agnostic experience. Inspired by this observation, this paper proposes a task-agnostic learning method (TAL for short) that can learn fragmented knowledge only from task-agnostic data to accomplish new tasks. TAL consists of four stages. First, the task-agnostic exploration is performed to collect data from interactions with the environment. The collected data is organized via a knowledge graph. Second, an action feature extractor is proposed and trained using the collected knowledge graph data for task-agnostic fragmented knowledge learning. Third, a candidate action generator is designed, which applies the action feature extractor on a new task to generate multiple candidate action sets. Finally, an action proposal network is designed to produce the probabilities for actions in a new task according to the environmental information. The probabilities are then used to generate order information for selecting actions to be executed from multiple candidate action sets to form the plan. Experiments on a virtual indoor scene show that the proposed method outperforms the state-of-the-art offline RL methods and IL methods by more than 20%.

Wei, Qinglai; Yuan, Beiming; Chen, Diancheng

doi: 10.48550/arxiv.2209.14946pmid: N/A

Abstract:This paper develops a new EiHi net to solve the out-of-distribution (OoD) generalization problem in deep learning. EiHi net is a model learning paradigm that can be blessed on any visual backbone. This paradigm can change the previous learning method of the deep model, namely find out correlations between inductive sample features and corresponding categories, which suffers from pseudo correlations between indecisive features and labels. We fuse SimCLR and VIC-Reg via explicitly and dynamically establishing the original - positive - negative sample pair as a minimal learning element, the deep model iteratively establishes a relationship close to the causal one between features and labels, while suppressing pseudo correlations. To further validate the proposed model, and strengthen the established causal relationships, we develop a human-in-the-loop strategy, with few guidance samples, to prune the representation space directly. Finally, it is shown that the developed EiHi net makes significant improvements in the most difficult and typical OoD dataset Nico, compared with the current SOTA results, without any domain ($e.g.$ background, irrelevant features) information.

Huertas-Tato, Javier; Martin, Alejandro; Camacho, David

doi: 10.48550/arxiv.2209.15373pmid: N/A

Abstract:Authors writing documents imprint identifying information within their texts: vocabulary, registry, punctuation, misspellings, or even emoji usage. Previous works use hand-crafted features or classification tasks to train their authorship models, leading to poor performance on out-of-domain authors. Using stylometric representations is more suitable, but this by itself is an open research challenge. In this paper, we propose PART, a contrastively trained model fit to learn \textbf{authorship embeddings} instead of semantics. We train our model on ~1.5M texts belonging to 1162 literature authors, 17287 blog posters and 135 corporate email accounts; a heterogeneous set with identifiable writing styles. We evaluate the model on current challenges, achieving competitive performance. We also evaluate our model on test splits of the datasets achieving zero-shot 72.39\% accuracy when bounded to 250 authors, a 54\% and 56\% higher than RoBERTa embeddings. We qualitatively assess the representations with different data visualizations on the available datasets, observing features such as gender, age, or occupation of the author.

Gergatsoulis, Manolis; Damigos, Matthew

doi: 10.48550/arxiv.2209.10272pmid: N/A

Abstract:In this paper, we investigate the problem of evaluating Basic Graph Patterns (BGP, for short, a subclass of SPARQL queries) over dynamic Linked Data graphs; i.e., Linked Data graphs that are continuously updated. We consider a setting where the updates are continuously received through a stream of messages and support both insertions and deletions of triples (updates are straightforwardly handled as a combination of deletions and insertions). In this context, we propose a set of in-memory algorithms minimizing the cached data to efficiently and continuously answer BGP queries. The queries are typically submitted into a system and continuously result in the delta answers while the update messages are processed. To efficiently and continuously evaluate the submitted query over the streaming data, as well as to minimize the amount of cached data, we propose an approach where the submitted query is decomposed into simpler subqueries and the query evaluation is achieved by combining the intermediate answers of the subqueries. Using this approach, the proposed algorithms compute the delta answers of a BGP query in polynomial time and space. Note that for certain subclasses of BGP queries, we show that the evaluation can be achieved in constant or linear time and space. Consolidating all the historical delta answers, the algorithms ensure that the answer to each query is constructed at any given time.

EK, Sannara; Portet, François; Lalanda, Philippe

doi: 10.1007/s00779-023-01776-3pmid: N/A

Abstract:Human Activity Recognition (HAR) on mobile devices has been demonstrated to be possible using neural models trained on data collected from the device's inertial measurement units. These models have used Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTMs), Transformers or a combination of these to achieve state-of-the-art results with real-time performance. However, these approaches have not been extensively evaluated in real-world situations where the input data may be different from the training data. This paper highlights the issue of data heterogeneity in machine learning applications and how it can hinder their deployment in pervasive settings. To address this problem, we propose and publicly release the code of two sensor-wise Transformer architectures called HART and MobileHART for Human Activity Recognition Transformer. Our experiments on several publicly available datasets show that these HART architectures outperform previous architectures with fewer floating point operations and parameters than conventional Transformers. The results also show they are more robust to changes in mobile position or device brand and hence better suited for the heterogeneous environments encountered in real-life settings. Finally, the source code has been made publicly available.

Chen, Kaiqi; Xiao, Junhao; Liu, Jialing; Tong, Qiyi; Zhang, Heng; Liu, Ruyu; Zhang, Jianhua; Ajoudani, Arash; Chen, Shengyong

doi: 10.48550/arxiv.2209.06428pmid: N/A

Abstract:Visual Simultaneous Localization and Mapping (vSLAM) has achieved great progress in the computer vision and robotics communities, and has been successfully used in many fields such as autonomous robot navigation and AR/VR. However, vSLAM cannot achieve good localization in dynamic and complex environments. Numerous publications have reported that, by combining with the semantic information with vSLAM, the semantic vSLAM systems have the capability of solving the above problems in recent years. Nevertheless, there is no comprehensive survey about semantic vSLAM. To fill the gap, this paper first reviews the development of semantic vSLAM, explicitly focusing on its strengths and differences. Secondly, we explore three main issues of semantic vSLAM: the extraction and association of semantic information, the application of semantic information, and the advantages of semantic vSLAM. Then, we collect and analyze the current state-of-the-art SLAM datasets which have been widely used in semantic vSLAM systems. Finally, we discuss future directions that will provide a blueprint for the future development of semantic vSLAM.

Mahdavian, Mohammad; Nikdel, Payam; TaherAhmadi, Mahdi; Chen, Mo

doi: 10.48550/arxiv.2209.07600pmid: N/A

Abstract:In this paper, we develop a neural network model to predict future human motion from an observed human motion history. We propose a non-autoregressive transformer architecture to leverage its parallel nature for easier training and fast, accurate predictions at test time. The proposed architecture divides human motion prediction into two parts: 1) the human trajectory, which is the hip joint 3D position over time and 2) the human pose which is the all other joints 3D positions over time with respect to a fixed hip joint. We propose to make the two predictions simultaneously, as the shared representation can improve the model performance. Therefore, the model consists of two sets of encoders and decoders. First, a multi-head attention module applied to encoder outputs improves human trajectory. Second, another multi-head self-attention module applied to encoder outputs concatenated with decoder outputs facilitates learning of temporal dependencies. Our model is well-suited for robotic applications in terms of test accuracy and speed, and compares favorably with respect to state-of-the-art methods. We demonstrate the real-world applicability of our work via the Robot Follow-Ahead task, a challenging yet practical case study for our proposed model.

Feldmann, Andreas Emil; Vu, Tung Anh

doi: 10.1007/s00453-025-01357-1pmid: N/A

Abstract:We consider generalizations of the $k$-Center problem in graphs of low doubling and highway dimension. For the Capacitated $k$-Supplier with Outliers (CkSwO) problem, we show an efficient parameterized approximation scheme (EPAS) when the parameters are $k$, the number of outliers and the doubling dimension of the supplier set. On the other hand, we show that for the Capacitated $k$-Center problem, which is a special case of CkSwO, obtaining a parameterized approximation scheme (PAS) is $\mathrm{W[1]}$-hard when the parameters are $k$, and the highway dimension. This is the first known example of a problem for which it is hard to obtain a PAS for highway dimension, while simultaneously admitting an EPAS for doubling dimension.