Graph Convolutional Networks (GCN) is a pioneering model for graph-based semi-supervised learning. However, GCN does not perform well on sparsely-labeled graphs. Its two-layer version cannot effectively propagate the label information to the whole graph structure (i.e., the under-smoothing problem) while its deep version over-smoothens and is hard to train (i.e., the over-smoothing problem). To solve these two issues, we propose a new graph neural network called GND-Nets (for Graph Neural Diffusion Networks) that exploits the local and global neighborhood information of a vertex in a single layer. Exploiting the shallow network mitigates the over-smoothing problem while exploiting the local and global neighborhood information mitigates the under-smoothing problem. The utilization of the local and global neighborhood information of a vertex is achieved by a new graph diffusion method called neural diffusions, which integrate neural networks into the conventional linear and nonlinear graph diffusions. The adoption of neural networks makes neural diffusions adaptable to different datasets. Extensive experiments on various sparsely-labeled graphs verify the effectiveness and efficiency of GND-Nets compared to state-of-the-art approaches.

AI-aided drug discovery (AIDD) is gaining increasing popularity due to its promise of making the search for new pharmaceuticals quicker, cheaper and more efficient. In spite of its extensive use in many fields, such as ADMET prediction, virtual screening, protein folding and generative chemistry, little has been explored in terms of the out-of-distribution (OOD) learning problem with \emph{noise}, which is inevitable in real world AIDD applications. In this work, we present DrugOOD, a systematic OOD dataset curator and benchmark for AI-aided drug discovery, which comes with an open-source Python package that fully automates the data curation and OOD benchmarking processes. We focus on one of the most crucial problems in AIDD: drug target binding affinity prediction, which involves both macromolecule (protein target) and small-molecule (drug compound). In contrast to only providing fixed datasets, DrugOOD offers automated dataset curator with user-friendly customization scripts, rich domain annotations aligned with biochemistry knowledge, realistic noise annotations and rigorous benchmarking of state-of-the-art OOD algorithms. Since the molecular data is often modeled as irregular graphs using graph neural network (GNN) backbones, DrugOOD also serves as a valuable testbed for \emph{graph OOD learning} problems. Extensive empirical studies have shown a significant performance gap between in-distribution and out-of-distribution experiments, which highlights the need to develop better schemes that can allow for OOD generalization under noise for AIDD.

Data available in scholarly knowledge graphs (SKGs) - i.e., "a graph of data intended to accumulate and convey knowledge of the real world, whose nodes represent entities of interest and whose edges represent potentially different relations between these entities" [14] - is growing continuously every day, leading to a plethora of challenges concerning, for instance, article exploration and visualization [17], article recommendation [3], citation recommendation [11], and Author Name Disambiguation (AND) [24], which is relevant for the purposes of the present article. In particular, AND refers to a specific task of entity resolution which aims at resolving author mentions in bibliographic references to real-world people. Author persistent identifiers, such as ORCIDs and VIAFs, simplify the AND activity since such identifiers can be used for reconciling entities defined as different objects and representing the same real-world person. However, the availability of such persistent identifiers in SKGs - such as OpenCitations (OC) [22], AMiner [27] and Microsoft Academic Knowledge Graph (MAKG) [10] - is characterized by very low coverage and, as such, additional and computationally-oriented techniques must be adopted to identify different authors as the same person. In the past, many automatic approaches have been developed to automatically address AND by using publications metadata (e.g., title, abstract, keywords, venue, affiliation, etc.) to extract some features which can be used in the disambiguation task. These methods vary widely from supervised learning methods to unsupervised learning including recently developed deep neural network-based architectures [31]. However, the existing SKGs do not provide all the relevant contextual information necessary to reuse effectively and efficiently such approaches, that often rely on pure textual data. In contrast with the approaches mentioned above, this study focuses on performing AND for scholarly data represented as linked data or included in SKGs by considering the multi-modal information available in such collections, i.e., the structural information consisting of entities and relations between them as well as text or numeric values associated with the authors and publications defined in the form of literals (family name, given name, publication title, venue title, year of publication, etc.). The proposed framework to address this task is named Literally Author Name Disambiguation (LAND), which focuses on tackling the following research questions: - Can Knowledge Graph Embeddings (KGEs) - i.e. a technique that enables the creation of a "dense representation of the graph in a continuous, low-dimensional vector space that can then be used for machine learning tasks"[13] - be used effectively for the downstream task of clustering, more specifically for author name disambiguation?

Because studied the joint resource allocation and edge device selection of the limited radio spectrum resource and increasing privacy to enhance learning performance. Both studies adopted the concerns, gathering geographically distributed data from a orthogonal multiple access (OMA) scheme, where the number large number of edge devices into a cloud server to enable of edge devices that can participate in each communication cloud artificial intelligence (AI) may not be practical. To this round is restricted by the number of available time/frequency end, edge AI has recently been envisioned as a promising AI resource blocks. The limited radio resource turns out to be the paradigm [1]. Unlike cloud AI that relies on a cloud server to main performance bottleneck of wireless FL. Fortunately, overthe-air conduct centralized training, edge AI exploits the computing computation (AirComp), as a non-orthogonal multiple power of multiple edge devices to perform model training access scheme, allows concurrent transmissions over the same with their own local data in a distributed manner. Federated radio channel to enable low-latency and spectrum-efficient learning (FL) [2], as a representative edge AI framework, wireless data aggregation [13]-[15], thereby mitigating the enables multiple edge devices to collaboratively train a shared communication bottleneck [16]. Motivated by this observation, model without exchanging their local data, which effectively various AirComp-assisted FL algorithms were proposed in alleviates the communication burden and privacy concerns.

We propose a novel statistical inference paradigm for zero-inflated multiway count data that dispenses with the need to distinguish between true and false zero counts. Our approach ignores all zero entries and applies zero-truncated Poisson regression on the positive counts. Inference is accomplished via tensor completion that imposes low-rank structure on the Poisson parameter space. Our main result shows that an $N$-way rank-$R$ parametric tensor $\boldsymbol{\mathscr{M}}\in(0,\infty)^{I\times \cdots\times I}$ generating Poisson observations can be accurately estimated from approximately $IR^2\log_2^2(I)$ non-zero counts for a nonnegative canonical polyadic decomposition. Several numerical experiments are presented demonstrating that our zero-truncated paradigm is comparable to the ideal scenario where the locations of false zero counts are known a priori.

We give tight statistical query (SQ) lower bounds for learnining halfspaces in the presence of Massart noise. In particular, suppose that all labels are corrupted with probability at most $\eta$. We show that for arbitrary $\eta \in [0,1/2]$ every SQ algorithm achieving misclassification error better than $\eta$ requires queries of superpolynomial accuracy or at least a superpolynomial number of queries. Further, this continues to hold even if the information-theoretically optimal error $\mathrm{OPT}$ is as small as $\exp\left(-\log^c(d)\right)$, where $d$ is the dimension and $0 < c < 1$ is an arbitrary absolute constant, and an overwhelming fraction of examples are noiseless. Our lower bound matches known polynomial time algorithms, which are also implementable in the SQ framework. Previously, such lower bounds only ruled out algorithms achieving error $\mathrm{OPT} + \epsilon$ or error better than $\Omega(\eta)$ or, if $\eta$ is close to $1/2$, error $\eta - o_\eta(1)$, where the term $o_\eta(1)$ is constant in $d$ but going to 0 for $\eta$ approaching $1/2$. As a consequence, we also show that achieving misclassification error better than $1/2$ in the $(A,\alpha)$-Tsybakov model is SQ-hard for $A$ constant and $\alpha$ bounded away from 1.

Recent years have seen an increased interest in the application of methods and techniques commonly associated with machine learning and artificial intelligence to spatial statistics. Here, in a celebration of the ten-year anniversary of the journal Spatial Statistics, we bring together normalizing flows, commonly used for density function estimation in machine learning, and spherical point processes, a topic of particular interest to the journal's readership, to present a new approach for modeling non-homogeneous Poisson process intensity functions on the sphere. The central idea of this framework is to build, and estimate, a flexible bijective map that transforms the underlying intensity function of interest on the sphere into a simpler, reference, intensity function, also on the sphere. Map estimation can be done efficiently using automatic differentiation and stochastic gradient descent, and uncertainty quantification can be done straightforwardly via nonparametric bootstrap. We investigate the viability of the proposed method in a simulation study, and illustrate its use in a proof-of-concept study where we model the intensity of cyclone events in the North Pacific Ocean. Our experiments reveal that normalizing flows present a flexible and straightforward way to model intensity functions on spheres, but that their potential to yield a good fit depends on the architecture of the bijective map, which can be difficult to establish in practice.

We consider the task of learning causal structures from data stored on multiple machines, and propose a novel structure learning method called distributed annealing on regularized likelihood score (DARLS) to solve this problem. We model causal structures by a directed acyclic graph that is parameterized with generalized linear models, so that our method is applicable to various types of data. To obtain a high-scoring causal graph, DARLS simulates an annealing process to search over the space of topological sorts, where the optimal graphical structure compatible with a sort is found by a distributed optimization method. This distributed optimization relies on multiple rounds of communication between local and central machines to estimate the optimal structure. We establish its convergence to a global optimizer of the overall score that is computed on all data across local machines. To the best of our knowledge, DARLS is the first distributed method for learning causal graphs with such theoretical guarantees. Through extensive simulation studies, DARLS has shown competing performance against existing methods on distributed data, and achieved comparable structure learning accuracy and test-data likelihood with competing methods applied to pooled data across all local machines. In a real-world application for modeling protein-DNA binding networks with distributed ChIP-Sequencing data, DARLS also exhibits higher predictive power than other methods, demonstrating a great advantage in estimating causal networks from distributed data.

Data heterogeneity is an intrinsic property of recommender systems, making models trained over the global data on the cloud, which is the mainstream in industry, non-optimal to each individual user's local data distribution. To deal with data heterogeneity, model personalization with on-device learning is a potential solution. However, on-device training using a user's small size of local samples will incur severe overfitting and undermine the model's generalization ability. In this work, we propose a new device-cloud collaborative learning framework, called CoDA, to break the dilemmas of purely cloud-based learning and on-device learning. The key principle of CoDA is to retrieve similar samples from the cloud's global pool to augment each user's local dataset to train the recommendation model. Specifically, after a coarse-grained sample matching on the cloud, a personalized sample classifier is further trained on each device for a fine-grained sample filtering, which can learn the boundary between the local data distribution and the outside data distribution. We also build an end-to-end pipeline to support the flows of data, model, computation, and control between the cloud and each device. We have deployed CoDA in a recommendation scenario of Mobile Taobao. Online A/B testing results show the remarkable performance improvement of CoDA over both cloud-based learning without model personalization and on-device training without data augmentation. Overhead testing on a real device demonstrates the computation, storage, and communication efficiency of the on-device tasks in CoDA.

We study the problem of semi-supervised learning with Graph Neural Networks (GNNs) in an active learning setup. We propose GraphPart, a novel partition-based active learning approach for GNNs. GraphPart first splits the graph into disjoint partitions and then selects representative nodes within each partition to query. The proposed method is motivated by a novel analysis of the classification error under realistic smoothness assumptions over the graph and the node features. Extensive experiments on multiple benchmark datasets demonstrate that the proposed method outperforms existing active learning methods for GNNs under a wide range of annotation budget constraints. In addition, the proposed method does not introduce additional hyperparameters, which is crucial for model training, especially in the active learning setting where a labeled validation set may not be available.