arXiv research

DEAL model predicts links for new nodes with only attribute info.

NPGNN improves graph link prediction by adapting to new graphs.

Paper proposes an inductive RGCN for few-shot link prediction in drug-repurposing.

In this paper, we study the problem of learning probabilistic logical rules for inductive and interpretable link prediction. Despite the importance of inductive link prediction, most previous works focused on transductive link prediction and cannot manage previously unseen entities. Moreover, they are black-box models …

Study shows gMPNNs struggle with OOD link prediction in larger test graphs.

Study links neural network inductive bias, feature learning, and generalization on Boolean functions.

The paper presents a new method to represent directed graphs using pseudo-Riemannian manifolds.

GEN tackles few-shot out-of-graph link prediction in evolving multi-relational graphs.

Link prediction aims to reveal missing edges in a graph. We address this task with a Gaussian process that is transformed using simplified graph convolutions to better leverage the inductive bias of the domain. To scale the Gaussian process model to large graphs, we introduce a variational inducing point method that pl…

A new DDPM for link prediction using sub-graph likelihood estimation.

New method for embedding large networks without attributes, achieving state-of-the-art performance.

Inductive representation learning on temporal graphs is an important step toward salable machine learning on real-world dynamic networks. The evolving nature of temporal dynamic graphs requires handling new nodes as well as capturing temporal patterns. The node embeddings, which are now functions of time, should repres…

The dominant paradigm for relation prediction in knowledge graphs involves learning and operating on latent representations (i.e., embeddings) of entities and relations. However, these embedding-based methods do not explicitly capture the compositional logical rules underlying the knowledge graph, and they are limited …

We apply Murasugi-Tristram inequality to real algebraic curves of odd degree on $RP^2$ with a deep nest, i.e. a nest of the depth $k-1$ where $2k+1$ is the degree. For such curves, the ingredients of the Murasugi-Tristram inequality can be computed (or estimated) inductively using the computations for iterated torus li…

Novel approach uses inductive biases for semiconductor etching.

We construct a new inductive basis of the Birman-Murakami-Wenzl algebra. Using it, we provide a new proof of the existence of the Markov trace on the BMW algebras affording the two-variable Kauffman polynomial. We prove also that all the transverse Markov traces on the BMW algebras are determined by the self-linking nu…

CAWs learn temporal network dynamics without node identities or edge attributes.

New proof shows exotic 4-manifolds exist without complex calculations.

OTI extends OTP for inductive semi-supervised learning.

Study shows different trajectory prediction models generalize better under OoD conditions.

How can we effectively encode evolving information over dynamic graphs into low-dimensional representations? In this paper, we propose DyRep, an inductive deep representation learning framework that learns a set of functions to efficiently produce low-dimensional node embeddings that evolves over time. The learned embe…

Study explores efficient data division for ICPs.

R2N learns interpretable rules and literals from numerical features.

This note introduces the method of cross-conformal prediction, which is a hybrid of the methods of inductive conformal prediction and cross-validation, and studies its validity and predictive efficiency empirically.

For an arbitrary positive integer $n$ and a pair $(p, q)$ of coprime integers, consider $n$ copies of a torus $(p,q)$ knot placed parallel to each other on the surface of the corresponding auxiliary torus: we call this assembly a torus $n$-link. We compute economical presentations of knot groups for torus links using t…

Explainability and effectiveness are two key aspects for building recommender systems. Prior efforts mostly focus on incorporating side information to achieve better recommendation performance. However, these methods have some weaknesses: (1) prediction of neural network-based embedding methods are hard to explain and …

In the article, the main ideas of the induction construction arXiv:1204.0194, arXiv:1110.4737, arXiv:1202.0941, arXiv:1208.4466 are considered in the form of sketches. The article establishes a link between Clifford algebras and alternative-elastic algebras at the level of connectors.

Models obtained by decision tree induction techniques excel in being interpretable.However, they can be prone to overfitting, which results in a low predictive performance. Ensemble techniques are able to achieve a higher accuracy. However, this comes at a cost of losing interpretability of the resulting model. This ma…

ICP provides interval predictions with high confidence coverage.

This is the last of a series of papers which give a necessary and sufficient condition for two essential simple loops on a 2-bridge sphere in a 2-bridge link complement to be homotopic in the link complement. The first paper of the series treated the case of the 2-bridge torus links, and the second paper treated the ca…

Transformers learn to use induction heads or shortcuts based on data diversity.

CNNs achieve remarkable performance by leveraging deep, over-parametrized architectures, trained on large datasets. However, they have limited generalization ability to data outside the training domain, and a lack of robustness to noise and adversarial attacks. By building better inductive biases, we can improve robust…

We give an alternative proof of that a critical knot of a Morse-Bott function $f: S^3 \rightarrow \mathbb{R}$ is a graph knot where the critical set of $f$ is a link in $S^3$. Our proof inducts on the number of index-1 critical knots of $f$.

New proof found for Khovanov's assertion about Frobenius algebra twists.

Network embedding has proved extremely useful in a variety of network analysis tasks such as node classification, link prediction, and network visualization. Almost all the existing network embedding methods learn to map the node IDs to their corresponding node embeddings. This design principle, however, hinders the ex…

Transformers tend to learn more symmetric functions in sequence data.

We propose an inductive matrix completion model without using side information. By factorizing the (rating) matrix into the product of low-dimensional latent embeddings of rows (users) and columns (items), a majority of existing matrix completion methods are transductive, since the learned embeddings cannot generalize …

Graphs (networks) are ubiquitous and allow us to model entities (nodes) and the dependencies (edges) between them. Learning a useful feature representation from graph data lies at the heart and success of many machine learning tasks such as classification, anomaly detection, link prediction, among many others. Many exi…

Document network embedding aims at learning representations for a structured text corpus i.e. when documents are linked to each other. Recent algorithms extend network embedding approaches by incorporating the text content associated with the nodes in their formulations. In most cases, it is hard to interpret the learn…

This work addresses encoding biases in neural networks by tailoring models with unsupervised losses.

Learning node embeddings that capture a node's position within the broader graph structure is crucial for many prediction tasks on graphs. However, existing Graph Neural Network (GNN) architectures have limited power in capturing the position/location of a given node with respect to all other nodes of the graph. Here w…

Machine learning refactors knowledge to improve learning efficiency.

Machine learning algorithms often contain many hyperparameters (HPs) whose values affect the predictive performance of the induced models in intricate ways. Due to the high number of possibilities for these HP configurations and their complex interactions, it is common to use optimization techniques to find settings th…

Methods that learn representations of nodes in a graph play a critical role in network analysis since they enable many downstream learning tasks. We propose Graph2Gauss - an approach that can efficiently learn versatile node embeddings on large scale (attributed) graphs that show strong performance on tasks such as lin…

Ensembles of classification and regression trees remain popular machine learning methods because they define flexible non-parametric models that predict well and are computationally efficient both during training and testing. During induction of decision trees one aims to find predicates that are maximally informative …

Proposes a neural framework to select subsets efficiently across different models.

Rule-based models are often used for data analysis as they combine interpretability with predictive power. We present RuleKit, a versatile tool for rule learning. Based on a sequential covering induction algorithm, it is suitable for classification, regression, and survival problems. The presence of a user-guided induc…

Soft geometric bias improves physical dynamics predictions.