{"total":10,"items":[{"citing_arxiv_id":"2605.22038","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Mixed Self-Exciting Process to Model Epileptic Seizures","primary_cat":"stat.ME","submitted_at":"2026-05-21T06:21:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A Bayesian mixed Hawkes process with Weibull baseline intensity and random effects is developed to model seizure clustering and heterogeneity in focal epilepsy from the Human Epilepsy Project data.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"functions for random effectsνi and ωi, respectively. Then the joint density function for complete datat and random effectsνandω, conditional on branching structureY , becomes p(t,ν,ω|Y,θ)∝ m∏ i=1  γϕ(νi)γξ(ωi)p(Ii|νi,Yi,θ) ni∏ y=1 p(Oiy|ωi,Yi,θ)   (8) ∝ m∏ i=1  γϕ(νi)γξ(ωi) exp{−ΛIi(Ti)} ∏ tij∈Ii λIi(tij) ni∏ y=1   exp{−ΛOiy(Ti)} ∏ tij∈Oiy λOiy(tij)     (9) 12 ∝ m∏ i=1   ϕϕ Γ(ϕ)νϕ−1 i exp(−ϕνi) ξξ Γ(ξ)ωξ−1 i exp(−ξωi) exp(−νiηiTi α) ∏ tij∈Ii νiηiαtα−1 ij × ni∏ y=1   exp { −ωiκi δ(1−exp{−δ(Ti−tiy)}) } ∏ tij∈Oiy ωiκi exp{−δ(tij−tiy)}    .(10) Rasmussen (2013) found that Bayesian sampling methods that condition on the cluster representation of a Hawkes process are more computationally efficient than those based on"},{"citing_arxiv_id":"2605.21307","ref_index":78,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"The Bayesian Gaussian Process Latent Variable Model for Spatio-Temporal Stream Networks","primary_cat":"stat.ME","submitted_at":"2026-05-20T15:35:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A variational inference-based framework for multi-output Gaussian process latent variable models on tails-up spatio-temporal stream networks using stream distance and process convolution.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20345","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Corrected Integrated Laplace Approximation for Bayesian Inference in Latent Gaussian Models","primary_cat":"stat.ML","submitted_at":"2026-05-19T18:02:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"An importance sampling correction is added to integrated Laplace approximation so that the approximate posterior for latent Gaussian models converges to the true posterior as the number of samples grows.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09577","ref_index":60,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Quadratic Forms in Gaussian Random Variables Theoretical Results and Applications","primary_cat":"eess.SP","submitted_at":"2026-05-10T14:40:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":1.0,"formal_verification":"none","one_line_summary":"A review summarizing definitions, canonical forms, exact and approximate distributions, numerical methods, applications, and open problems for quadratic forms in real and complex Gaussian variables, including multiforms and ratios.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"E[Q1Q2] = 2tr(A1ΣA 2Σ) + 4µTA1ΣA 2µ + [ µTA1µ+tr(A1Σ) ] [ µTA2µ+tr(A2Σ) ] , (4.6) and Cov(Q1,Q 2) = 2tr(A1ΣA 2Σ) + 4µTA1ΣA 2µ.(4.7) 79 Quadratic Forms in Gaussian Random VariablesPreprint For two complete complex forms, E[Q1Q2] =tr(A 1ΣA 2Σ) +µHA1ΣA 2µ+µHA2ΣA 1µ + [ µHA1µ+tr(A1Σ) ] [ µHA2µ+tr(A2Σ) ] , (4.8) and Cov(Q1,Q 2) =tr(A 1ΣA 2Σ) +µHA1ΣA 2µ+µHA2ΣA 1µ.(4.9) Ghazal [60] derives a recurrence formula that evaluates the expectation of products of arbitrary number of quadratic forms in the formxTAix,i= 1,2,...M, wherex∼N(0,Σ). 4.2 Independence An obvious approach toward simplifying multiforms is the independence assumption. Independence reduces product moments, cdf and pdf of multiforms to a set of single-form problems, respectively, as follows"},{"citing_arxiv_id":"2605.09562","ref_index":46,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Laplace Variational Inference for Dirichlet Process Mixtures of Marked Poisson Point Processes","primary_cat":"stat.ME","submitted_at":"2026-05-10T14:25:32+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A Dirichlet process mixture model for marked Poisson point processes with squared-link intensities and Laplace variational inference jointly infers clusters, cluster count, and continuous mark-specific intensity surfaces.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"to conjugate parametric families and includes an explicit within-component cardinality model. As a result, clustering can be driven jointly by set size and simple feature distributions, which is not aligned with our goal of normalizing out subject-specific event volume and learning smooth continuous mark- dependent intensity surfaces. The closest structural analogue is Yin et al.[46], who clustered repeated marked event sequences through a mixture of multi-level marked LGCPs, but their method is tailored to matrix-valued repeated observations, point-estimate centered, and requires selecting the number of clusters and a smoothing bandwidth. Hawkes- and neural-temporal-point-process clustering methods target excitation or neural sequence dynamics instead [43, 47, 10]."},{"citing_arxiv_id":"2605.06742","ref_index":102,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Bayesian Modeling and Prediction of Generalized Contact Matrices","primary_cat":"stat.ME","submitted_at":"2026-05-07T14:30:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A Bayesian model for multi-feature contact matrices that uses tensor structures and contingency table theory to satisfy structural constraints and impute missing contact features, validated on simulations and US/German survey data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.19972","ref_index":21,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Principal Nested Cones","primary_cat":"stat.ME","submitted_at":"2026-04-21T20:28:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Principal Nested Cones is a nonlinear dimension reduction technique that projects cone-structured data onto nested lower-dimensional cones to jointly represent size and shape variation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.21807","ref_index":5,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MIBoost: A gradient boosting algorithm for variable selection after multiple imputation","primary_cat":"stat.ML","submitted_at":"2025-07-29T13:42:38+00:00","verdict":"CONDITIONAL","verdict_confidence":"MODERATE","novelty_score":7.0,"formal_verification":"none","one_line_summary":"MIBoost extends gradient boosting to multiple imputation by defining a single loss function that produces one set of selected variables across all imputed datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2408.15701","ref_index":15,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Robust discriminant analysis","primary_cat":"stat.ME","submitted_at":"2024-08-28T10:59:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A review paper that identifies the outlier sensitivity of classical discriminant analysis and summarizes robust versions based on resistant location and scatter estimators plus diagnostic graphics.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2406.04098","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Large-Scale Neutral Comparison Study of Survival Models on Low-Dimensional Data","primary_cat":"stat.ML","submitted_at":"2024-06-06T14:13:38+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Large-scale neutral benchmark of survival models on low-dimensional right-censored data finds Cox PH performs comparably to more complex methods across discrimination, calibration, and predictive metrics.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"Forest (RFSRC) [48]; x) Random Survival Forest (RAN) [48, 105]; xi) Conditional Inference Forest (CIF) [46]; xii) Oblique Random Survival Forest (ORSF) [49]; xiii) Relative Risk Tree (RRT) [17]; xiv) Model-Based Boosting (MBST) [20]; xv) CoxBoost (CoxB) [12]; xvi) XGBoost with Cox objec- tive (XGBCox) [25]; xvii) XGBoost with AFT objective (XGBAFT) [7]; and xviii) SSVM-Hybrid (SSVM) [97]. The full table of all models including respective software packages and versions is given in Appendix C. In our selection we focused on well-established models with robust implementations, provided as well maintained packages or wrapper functions within benchmarking software. This excludes some recently proposed DL based methods like DeepSurv [55] and DeepHit [68], which have"}],"limit":50,"offset":0}