Unsupervised Graph Domain Adaptation (UGDA) has emerged as a practical solution to transfer knowledge from a label-rich source graph to a completely unlabelled target graph. However, most methods require a labelled source graph to provide supervision signals, which might not be accessible in the real-world settings due to regulations and privacy concerns. In this paper, we explore the scenario of source-free unsupervised graph domain adaptation, which tries to address the domain adaptation problem without accessing the labelled source graph. Specifically, we present a novel paradigm called GraphCTA, which performs model adaptation and graph adaptation collaboratively through a series of procedures: (1) conduct model adaptation based on node's neighborhood predictions in target graph considering both local and global information; (2) perform graph adaptation by updating graph structure and node attributes via neighborhood contrastive learning; and (3) the updated graph serves as an input to facilitate the subsequent iteration of model adaptation, thereby establishing a collaborative loop between model adaptation and graph adaptation. Comprehensive experiments are conducted on various public datasets. The experimental results demonstrate that our proposed model outperforms recent source-free baselines by large margins.

This paper presents a novel hierarchical federated learning algorithm within multiple sets that incorporates quantization for communication-efficiency and demonstrates resilience to statistical heterogeneity. Unlike conventional hierarchical federated learning algorithms, our approach combines gradient aggregation in intra-set iterations with model aggregation in inter-set iterations. We offer a comprehensive analytical framework to evaluate its optimality gap and convergence rate, comparing these aspects with those of conventional algorithms. Additionally, we develop a problem formulation to derive optimal system parameters in a closed-form solution. Our findings reveal that our algorithm consistently achieves high learning accuracy over a range of parameters and significantly outperforms other hierarchical algorithms, particularly in scenarios with heterogeneous data distributions.

We introduce a notion of distance between supervised learning problems, which we call the Risk distance. This optimal-transport-inspired distance facilitates stability results; one can quantify how seriously issues like sampling bias, noise, limited data, and approximations might change a given problem by bounding how much these modifications can move the problem under the Risk distance. With the distance established, we explore the geometry of the resulting space of supervised learning problems, providing explicit geodesics and proving that the set of classification problems is dense in a larger class of problems. We also provide two variants of the Risk distance: one that incorporates specified weights on a problem's predictors, and one that is more sensitive to the contours of a problem's risk landscape.

We investigate the replay buffer in rehearsal-based approaches for graph continual learning (GCL) methods. Existing rehearsal-based GCL methods select the most representative nodes for each class and store them in a replay buffer for later use in training subsequent tasks. However, we discovered that considering only the class representativeness of each replayed node makes the replayed nodes to be concentrated around the center of each class, incurring a potential risk of overfitting to nodes residing in those regions, which aggravates catastrophic forgetting. Moreover, as the rehearsal-based approach heavily relies on a few replayed nodes to retain knowledge obtained from previous tasks, involving the replayed nodes that have irrelevant neighbors in the model training may have a significant detrimental impact on model performance. In this paper, we propose a GCL model named DSLR, specifically, we devise a coverage-based diversity (CD) approach to consider both the class representativeness and the diversity within each class of the replayed nodes. Moreover, we adopt graph structure learning (GSL) to ensure that the replayed nodes are connected to truly informative neighbors. Extensive experimental results demonstrate the effectiveness and efficiency of DSLR. Our source code is available at https://github.com/seungyoon-Choi/DSLR_official.

This work studies the general principles of improving the learning of language models (LMs), which aims at reducing the necessary training steps for achieving superior performance. Specifically, we present a theory for the optimal learning of LMs. We first propose an objective that optimizes LM learning by maximizing the data compression ratio in an "LM-training-as-lossless-compression" view. Then, we derive a theorem, named Learning Law, to reveal the properties of the dynamics in the optimal learning process under our objective. The theorem is then validated by experiments on a linear classification and a real-world language modeling task. Finally, we empirically verify that the optimal learning of LMs essentially stems from the improvement of the coefficients in the scaling law of LMs, indicating great promise and significance for designing practical learning acceleration methods. Our code can be found at https://aka.ms/LearningLaw.

Precise estimation of cardiac patients' current and future comorbidities is an important factor in prioritizing continuous physiological monitoring and new therapies. ML models have shown satisfactory performance in short-term mortality prediction of patients with heart disease, while their utility in long-term predictions is limited. This study aims to investigate the performance of tree-based ML models on long-term mortality prediction and the effect of two recently introduced biomarkers on long-term mortality. This study utilized publicly available data from CCHIA at the Ministry of Health and Welfare, Taiwan, China. Medical records were used to gather demographic and clinical data, including age, gender, BMI, percutaneous coronary intervention (PCI) status, and comorbidities such as hypertension, dyslipidemia, ST-segment elevation myocardial infarction (STEMI), and non-STEMI. Using medical and demographic records as well as two recently introduced biomarkers, brachial pre-ejection period (bPEP) and brachial ejection time (bET), collected from 139 patients with acute myocardial infarction, we investigated the performance of advanced ensemble tree-based ML algorithms (random forest, AdaBoost, and XGBoost) to predict all-cause mortality within 14 years. The developed ML models achieved significantly better performance compared to the baseline LR (C-Statistic, 0.80 for random forest, 0.79 for AdaBoost, and 0.78 for XGBoost, vs 0.77 for LR) (P-RF<0.001, PAdaBoost<0.001, PXGBoost<0.05). Adding bPEP and bET to our feature set significantly improved the algorithms' performance, leading to an absolute increase in C-Statistic of up to 0.03 (C-Statistic, 0.83 for random forest, 0.82 for AdaBoost, and 0.80 for XGBoost, vs 0.74 for LR) (P-RF<0.001, PAdaBoost<0.001, PXGBoost<0.05). This advancement may enable better treatment prioritization for high-risk individuals.

Over a significant period in the past, the convolutional network [He et al., 2016; Zheng et al., 2014; Middlehurst This paper introduces ConvTimeNet, a novel deep et al., 2023] has played a crucial role in time series analysis, hierarchical fully convolutional network designed largely due to its inherent properties that strike an excellent to serve as a general-purpose model for time series balance between computational efficiency and representation analysis. The key design of this network is twofold, quality. Data back to the past years, many representative designed to overcome the limitations of traditional works [Bagnall et al., 2017] of time series analysis typically convolutional networks. Firstly, we propose an employ convolutional networks as the backbone. For adaptive segmentation of time series into sub-series instance, temporal convolutional network (TCN[Bai et al., level patches, treating these as fundamental modeling 2018]) and its variants are widely used in modeling temporal units. This setting avoids the sparsity semantics variation dependence for the time series forecasting task. Furthermore, associated with raw point-level time steps. Secondly, a large number of works (such as InceptionTime[Ismail we design a fully convolutional block by Fawaz et al., 2020], MiniRocket[Dempster et al., 2021], skillfully integrating deepwise and pointwise convolution and MCNN[Cui et al., 2016]) are also proposed by employing operations, following the advanced building convolutional networks to identify informative patterns block style employed in Transformer encoders.

Point cloud segmentation, which helps us understand the environment of specific structures and objects, can be performed in class-specific and class-agnostic ways. We propose a novel region-based transformer model called Region-Transformer for performing class-agnostic point cloud segmentation. The model utilizes a region-growth approach and self-attention mechanism to iteratively expand or contract a region by adding or removing points. It is trained on simulated point clouds with instance labels only, avoiding semantic labels. Attention-based networks have succeeded in many previous methods of performing point cloud segmentation. However, a region-growth approach with attention-based networks has yet to be used to explore its performance gain. To our knowledge, we are the first to use a self-attention mechanism in a region-growth approach. With the introduction of self-attention to region-growth that can utilize local contextual information of neighborhood points, our experiments demonstrate that the Region-Transformer model outperforms previous class-agnostic and class-specific methods on indoor datasets regarding clustering metrics. The model generalizes well to large-scale scenes. Key advantages include capturing long-range dependencies through self-attention, avoiding the need for semantic labels during training, and applicability to a variable number of objects. The Region-Transformer model represents a promising approach for flexible point cloud segmentation with applications in robotics, digital twinning, and autonomous vehicles.

We examined the use of the Ising model as an $L_0$ regularization method for field-aware factorization machines (FFM). This approach improves generalization performance and has the advantage of simultaneously determining the best feature combinations for each of several groups. We can deepen the interpretation and understanding of the model from the similarities and differences in the features selected in each group.

It is observed empirically that the large language models (LLM), trained with a variant of regression loss using numerous corpus from the Internet, can unveil causal associations to some extent. This is contrary to the traditional wisdom that ``association is not causation'' and the paradigm of traditional causal inference in which prior causal knowledge should be carefully incorporated into the design of methods. It is a mystery why causality, in a higher layer of understanding, can emerge from the regression task that pursues associations. In this paper, we claim the emergence of causality from association-oriented training can be attributed to the coupling effects from the heterogeneity of the source data, stochasticity of training algorithms, and over-parameterization of the learning models. We illustrate such an intuition using a simple but insightful model that learns invariance, a quasi-causality, using regression loss. To be specific, we consider multi-environment low-rank matrix sensing problems where the unknown r-rank ground-truth d*d matrices diverge across the environments but contain a lower-rank invariant, causal part. In this case, running pooled gradient descent will result in biased solutions that only learn associations in general. We show that running large-batch Stochastic Gradient Descent, whose each batch being linear measurement samples randomly selected from a certain environment, can successfully drive the solution towards the invariant, causal solution under certain conditions. This step is related to the relatively strong heterogeneity of the environments, the large step size and noises in the optimization algorithm, and the over-parameterization of the model. In summary, we unveil another implicit bias that is a result of the symbiosis between the heterogeneity of data and modern algorithms, which is, to the best of our knowledge, first in the literature.