Variational learning of individual survival distributions

Xiu, Zidi; Tao, Chenyang; Henao, Ricardo

doi:10.1145/3368555.3384454

Cited by 10 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study showed that a high FLC was significantly predictive of worse overall survival. Survivability predictions on this data set tend to give a C-index between 75% and 80% [11].…”

Section: Datasets Descriptionmentioning

confidence: 84%

“…Variational inference is still an underexplored area in Survival Analysis. In a recent work, Xiu et al introduced the Variational Survival Inference (VSI) model [11], which uses variational inference to approximate p(t|x). The VSI model uses two encoders p(z|x) and q(z|x, t) and pushes these two distributions to be similar, using Kullback-Leibler divergence, to allow the model to better account for interactions between covariates and event times.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Concordance Index decomposition: a measure for a deeper understanding of survival prediction models

Alabdallah¹,

Ohlsson²,

Pashami³

et al. 2022

Preprint

View full text Add to dashboard Cite

The Concordance Index (C-index) is a commonly used metric in Survival Analysis to evaluate how good a prediction model is. This paper proposes a decomposition of the C-Index into a weighted harmonic mean of two quantities: one for ranking observed events versus other observed events, and the other for ranking observed events versus censored cases. This decomposition allows a more fine-grained analysis of the pros and cons of survival prediction methods. The utility of the decomposition is demonstrated using three benchmark survival analysis models (Cox Proportional Hazard, Random Survival Forest, and Deep Adversarial Time-to-Event Network) together with a new variational generative neuralnetwork-based method (SurVED), which is also proposed in this paper. The demonstration is done on four publicly available datasets with varying censoring levels.The analysis with the C-index decomposition shows that all methods essentially perform equally well when the censoring level is high because of the dominance of the term measuring the ranking of events versus censored cases. In contrast, some methods deteriorate when the censoring level decreases because they do not rank the events versus other events well.

show abstract

“…The study showed that a high FLC was significantly predictive of worse overall survival. Survivability predictions on this data set tend to give a C-index between 75% and 80% [11].…”

Section: Datasets Descriptionmentioning

confidence: 84%

Section: Introductionmentioning

confidence: 99%

The Concordance Index decomposition: a measure for a deeper understanding of survival prediction models

Alabdallah¹,

Ohlsson²,

Pashami³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Unlike standard observational studies with well-defined and observable outcomes, time-to-event measures fall into the range of survival analyses, which focus on the length of time until the occurrence of a well-defined outcome [23,33]. A characteristic feature in the study of time-to-event distributions is the presence of censored instances: events that do not occur during the follow-up period of a subject.…”

Section: Time To Next Engagementmentioning

confidence: 99%

Feedback Effect in User Interaction with Intelligent Assistants: Delayed Engagement, Adaption and Drop-out

Xiu¹,

Cheng²,

Sun³

et al. 2023

Preprint

View full text Add to dashboard Cite

With the growing popularity of intelligent assistants (IAs), evaluating IA quality becomes an increasingly active field of research. This paper identifies and quantifies the feedback effect, a novel component in IA-user interactions -how the capabilities and limitations of the IA influence user behavior over time. First, we demonstrate that unhelpful responses from the IA cause users to delay or reduce subsequent interactions in the short term via an observational study. Next, we expand the time horizon to examine behavior changes and show that as users discover the limitations of the IA's understanding and functional capabilities, they learn to adjust the scope and wording of their requests to increase the likelihood of receiving a helpful response from the IA. Our findings highlight the impact of the feedback effect at both the micro and meso levels. We further discuss its macro-level consequences: unsatisfactory interactions continuously reduce the likelihood and diversity of future user engagements in a feedback loop.

show abstract

“…More recent papers have explored the use of Discrete time models (Lee et al, 2018), recurrent neural architectures (Lee et al, 2019a) as well as fully parametric methods (Nagpal et al, 2020) for modelling survival outcomes in the presence of censoring. More involved techniques have involved the use of ensembles with black box optimization, auto encoding variational bayes (Chapfuwa et al, 2020;Xiu et al, 2020), as well as adversarial methods (Chapfuwa et al, 2018) to estimate survival outcomes.…”

Section: Related Workmentioning

confidence: 99%

Deep Cox Mixtures for Survival Regression

Nagpal¹,

Yadlowsky²,

Rostamzadeh³

et al. 2021

Preprint

View full text Add to dashboard Cite

Survival analysis is a challenging variation of regression modeling because of the presence of censoring, where the outcome measurement is only partially known, due to, for example, loss to follow up. Such problems come up frequently in medical applications, making survival analysis a key endeavor in biostatistics and machine learning for healthcare, with Cox regression models being amongst the most commonly employed models. We describe a new approach for survival analysis regression models, based on learning mixtures of Cox regressions to model individual survival distributions. We propose an approximation to the Expectation Maximization algorithm for this model that does hard assignments to mixture groups to make optimization efficient. In each group assignment, we fit the hazard ratios within each group using deep neural networks, and the baseline hazard for each mixture component non-parametrically.We perform experiments on multiple real world datasets, and look at the mortality rates of patients across ethnicity and gender. We emphasize the importance of calibration in healthcare settings and demonstrate that our approach outperforms classical and modern survival analysis baselines, both in terms of discriminative performance and calibration, with large gains in performance on the minority demographics.

show abstract

Variational learning of individual survival distributions

Cited by 10 publications

References 35 publications

The Concordance Index decomposition: a measure for a deeper understanding of survival prediction models

The Concordance Index decomposition: a measure for a deeper understanding of survival prediction models

Feedback Effect in User Interaction with Intelligent Assistants: Delayed Engagement, Adaption and Drop-out

Deep Cox Mixtures for Survival Regression

Contact Info

Product

Resources

About