Multivariate time series data is ubiquitous, and forecasting it has important applications in many domains. However, its complex spatial dependencies and non-linear temporal dynamics can be challenging for traditional techniques. Existing methods tackle these challenges by learning the two dimensions separately. Here, we introduce Temporal Graphormer (T-Graphormer), a Transformer-based approach capable of modelling spatiotemporal correlations simultaneously. By incorporating temporal dynamics in the Graphormer architecture, each node attends to all other nodes within the graph sequence. Our design enables the model to capture rich spatiotemporal patterns with minimal reliance on predefined spacetime inductive biases. We validate the effectiveness of T-Graphormer on real-world traffic prediction benchmark datasets. Compared to state-of-the-art methods, T-Graphormer reduces root mean squared error (RMSE) and mean absolute percentage error (MAPE) by up to 10%.

In this work, we explore the capabilities of multiplexed gradient descent (MGD), a scalable and efficient perturbative zeroth-order training method for estimating the gradient of a loss function in hardware and training it via stochastic gradient descent. We extend the framework to include both weight and node perturbation, and discuss the advantages and disadvantages of each approach. We investigate the time to train networks using MGD as a function of network size and task complexity. Previous research has suggested that perturbative training methods do not scale well to large problems, since in these methods the time to estimate the gradient scales linearly with the number of network parameters. However, in this work we show that the time to reach a target accuracy--that is, actually solve the problem of interest--does not follow this undesirable linear scaling, and in fact often decreases with network size. Furthermore, we demonstrate that MGD can be used to calculate a drop-in replacement for the gradient in stochastic gradient descent, and therefore optimization accelerators such as momentum can be used alongside MGD, ensuring compatibility with existing machine learning practices. Our results indicate that MGD can efficiently train large networks on hardware, achieving accuracy comparable to backpropagation, thus presenting a practical solution for future neuromorphic computing systems.

Mediterranean cyclones are extreme meteorological events of which much less is known compared to their tropical, oceanic counterparts. The raising interest in such phenomena is due to their impact on a region increasingly more affected by climate change, but a precise characterization remains a non trivial task. In this work we showcase how a Bayesian algorithm (Latent Dirichlet Allocation) can classify Mediterranean cyclones relying on wind velocity data, leading to a drastic dimensional reduction that allows the use of supervised statistical learning techniques for detecting and tracking new cyclones.

Source-free domain adaptation in visual emotion recognition (SFDA-VER) is a highly challenging task that requires adapting VER models to the target domain without relying on source data, which is of great significance for data privacy protection. However, due to the unignorable disparities between visual emotion data and traditional image classification data, existing SFDA methods perform poorly on this task. In this paper, we investigate the SFDA-VER task from a fuzzy perspective and identify two key issues: fuzzy emotion labels and fuzzy pseudo-labels. These issues arise from the inherent uncertainty of emotion annotations and the potential mispredictions in pseudo-labels. To address these issues, we propose a novel fuzzy-aware loss (FAL) to enable the VER model to better learn and adapt to new domains under fuzzy labels. Specifically, FAL modifies the standard cross entropy loss and focuses on adjusting the losses of non-predicted categories, which prevents a large number of uncertain or incorrect predictions from overwhelming the VER model during adaptation. In addition, we provide a theoretical analysis of FAL and prove its robustness in handling the noise in generated pseudo-labels. Extensive experiments on 26 domain adaptation sub-tasks across three benchmark datasets demonstrate the effectiveness of our method.

Graph distillation (GD) is an effective approach to extract useful information from large-scale network structures. However, existing methods, which operate in Euclidean space to generate condensed graphs, struggle to capture the inherent tree-like geometry of real-world networks, resulting in distilled graphs with limited task-specific information for downstream tasks. Furthermore, these methods often fail to extract dynamic properties from graphs, which are crucial for understanding information flow and facilitating graph continual learning. This paper presents the Hyperbolic Graph Distillation with Random Walks Optimization (HyDRO), a novel graph distillation approach that leverages hyperbolic embeddings to capture complex geometric patterns and optimize the spectral gap in hyperbolic space. Experiments show that HyDRO demonstrates strong task generalization, consistently outperforming state-of-the-art methods in both node classification and link prediction tasks. HyDRO also effectively preserves graph random walk properties, producing condensed graphs that achieve enhanced performance in continual graph learning. Additionally, HyDRO achieves competitive results on mainstream graph distillation benchmarks, while maintaining a strong balance between privacy and utility, and exhibiting robust resistance to noises.

As probabilistic models continue to permeate various facets of our society and contribute to scientific advancements, it becomes a necessity to go beyond traditional metrics such as predictive accuracy and error rates and assess their trustworthiness. Grounded in the competence-based theory of trust, this work formalizes I-trustworthy framework -- a novel framework for assessing the trustworthiness of probabilistic classifiers for inference tasks by linking local calibration to trustworthiness. To assess I-trustworthiness, we use the local calibration error (LCE) and develop a method of hypothesis-testing. This method utilizes a kernel-based test statistic, Kernel Local Calibration Error (KLCE), to test local calibration of a probabilistic classifier. This study provides theoretical guarantees by offering convergence bounds for an unbiased estimator of KLCE. Additionally, we present a diagnostic tool designed to identify and measure biases in cases of miscalibration. The effectiveness of the proposed test statistic is demonstrated through its application to both simulated and real-world datasets. Finally, LCE of related recalibration methods is studied, and we provide evidence of insufficiency of existing methods to achieve I-trustworthiness.

Model fingerprinting is a widely adopted approach to safeguard the intellectual property rights of open-source models by preventing their unauthorized reuse. It is promising and convenient since it does not necessitate modifying the protected model. In this paper, we revisit existing fingerprinting methods and reveal that they are vulnerable to false claim attacks where adversaries falsely assert ownership of any third-party model. We demonstrate that this vulnerability mostly stems from their untargeted nature, where they generally compare the outputs of given samples on different models instead of the similarities to specific references. Motivated by these findings, we propose a targeted fingerprinting paradigm (i.e., FIT-Print) to counteract false claim attacks. Specifically, FIT-Print transforms the fingerprint into a targeted signature via optimization. Building on the principles of FIT-Print, we develop bit-wise and list-wise black-box model fingerprinting methods, i.e., FIT-ModelDiff and FIT-LIME, which exploit the distance between model outputs and the feature attribution of specific samples as the fingerprint, respectively. Extensive experiments on benchmark models and datasets verify the effectiveness, conferrability, and resistance to false claim attacks of our FIT-Print.

School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China Emails: {23b903088, zhangzhaoxin, 22s030153, li.ning, 22b303010}@stu.hit.edu.cn School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China Email: xinran_Liu@bupt.edu.cn Abstract --Fine-grained IP geolocation plays a critical role in applications such as location-based services and cybersecurity. Most existing fine-grained IP geolocation methods are regression-based; however, due to noise in the input data, these methods typically encounter kilometer-level prediction errors and provide incorrect region information for users. T o address this issue, this paper proposes a novel hierarchical multi-label classification framework for IP region prediction, named HMCGeo. This framework treats IP geolocation as a hierarchical multi-label classification problem and employs residual connection-based feature extraction and attention prediction units to predict the target host region across multiple geographical granularities. Furthermore, we introduce probabilistic classification loss during training, combining it with hierarchical cross-entropy loss to form a composite loss function. IP region prediction experiments on the New Y ork, Los Angeles, and Shanghai datasets demonstrate that HMCGeo achieves superior performance across all geographical granularities, significantly outperforming existing IP geolocation methods. P geolocation is a technique used to predict the geographical location of a host based on its IP address [1], playing a crucial role in location-based services, network topology optimization, and cybersecurity [2], [3], [4], [5], [6], [7], [8]. Using IP geolocation technology, online services and applications infer the geographical location of users to deliver localized weather updates, news, and event notifications [3]. Internet service providers (ISPs) estimate the approximate location of target hosts to optimize traffic transmission paths, reduce network latency, and improve transmission efficiency [4]. Network analysts examine the geographical origins of incoming traffic to assess security threats from suspicious addresses. This research was supported by the National Key R&D Program of China (2024QY1103, 2018YFB18002). Based on the accuracy of prediction results, IP geolocation is categorized into coarse-grained and fine-grained geolocation. Coarse-grained IP geolocation predicts the location of a target host by utilizing allocation information such as Autonomous System Numbers (ASN), ISP, and BGP, or by analyzing the relationship between latency and distance. These methods construct geolocation databases that provide location information at the country or city level. Building on this foundation, fine-grained IP geolocation reduces prediction errors to a few kilometers in certain regions by leveraging richer landmarks or employing more effective prediction methods.

Weaknesses: At Eq. 5, the authors introduce two sampling based estimators of the lower bound (LB). I am not sure why the authors introduced both as an estimator for the LB: The second estimator is an unbiased estimator of the (log) marginal likelihood (ML). Though it could technically be considered a biased estimator of LB, I do not see why it should be introduced as such, since it is the unbiased estimator of the exact value the authors are hoping to approximate. Actually, in the following sentence the authors write that the second estimator's bias decreases as J is increased, which is very much expected, if not almost trivial considering the point above. Another point is that when J 1 the two estimators are algebraically the same, therefore the first one also becomes (a noisy) unbiased estimator of ML.

The work considers SGD training for Bayesian linear models and illustrate a connection between training speed and generalization and why SGD tends to select simpler models. In particular, the work illustrates that a particular type of posterior sampling from gradient descent yields same model rankings as that based on the true posterior under suitable assumptions. Experiments on deep nets are also presented. The reviewers liked the work overall, but felt that some aspects of the exposition were unclear, the transition and implications for deep nets is not quite convincing especially since there is now better understanding of both optimization and generalization in deep nets, and baseline comparisons (e.g., sgld, L2 regularization, dropout, etc.) would strengthen the work.