If you tell a learning model that you prefer an alternative a over another alternative b, then you probably expect the model to be monotone, that is, the valuation of a increases, and that of bdecreases. Yet, perhaps surprisingly, many widely deployed comparison-based preference learning models, including large language models, fail to have this guarantee. Until now, the only comparison-based preference learning algorithms that were proved to be monotone are the Generalized BradleyTerry models [10]. Yet, these models are unable to generalize to uncompared data. In this paper, we advance the understanding of the set of models with generalization ability that are monotone. Namely, we propose a new class of Linear Generalized Bradley-Terry models with Diffusion Priors, and identify sufficient conditions on alternatives' embeddings that guarantee monotonicity. Our experiments show that this monotonicity is far from being a general guarantee, and that our new class of generalizing models improves accuracy, especially when the dataset is limited.

Neural Information Processing Systems http://nips.cc/

Several graph data mining, signal processing, and machine learning downstream tasks rely on information related to the eigenvectors of the associated adjacency or Laplacian matrix. Classical eigendecomposition methods are powerful when the matrix remains static but cannot be applied to problems where the matrix entries are updated or the number of rows and columns increases frequently. Such scenarios occur routinely in graph analytics when the graph is changing dynamically and either edges and/or nodes are being added and removed. This paper puts forth a new algorithmic framework to update the eigenvectors associated with the leading eigenvalues of an initial adjacency or Laplacian matrix as the graph evolves dynamically. The proposed algorithm is based on Rayleigh-Ritz projections, in which the original eigenvalue problem is projected onto a restricted subspace which ideally encapsulates the invariant subspace associated with the sought eigenvectors. Following ideas from eigenvector perturbation analysis, we present a new methodology to build the projection subspace. The proposed framework features lower computational and memory complexity with respect to competitive alternatives while empirical results show strong qualitative performance, both in terms of eigenvector approximation and accuracy of downstream learning tasks of central node identification and node clustering.

The main result is presented in Theorem 2. According to the definition of the Fiedler vector, we have ( L + L)( f + f) = ( λ + λ)( f + f). We outline the proof below for interested readers. The main result is presented in Theorem 2. We first present Stewart's theorem in Lemma 1 to assist Actual times may differ depending on hardware and environment. We also show the number of model parameters required for each method in Table S3. Hyper-parameters were selected based on a coarse search on the validation set.

Determinantal point processes (DPPs) are well-suited to recommender systems where the goal is to generate collections of diverse, high-quality items.

Then,itsamplesafixednumber of nodes by the calculated probability,and recursively conducts such procedure per layer toconstruct the whole computation graph.

Learning a graph with a specific structure is essential for interpretability and identification of the relationships among data. It is well known that structured graph learning from observedsamples isanNP-hard combinatorial problem. In this paper, we first show that for a set of important graph families it is possible toconvertthestructural constraints ofstructure intoeigenvalueconstraints ofthe graph Laplacianmatrix.

Neural Information Processing Systems http://nips.cc/