• Corpus ID: 239050076

Towards modelling hazard factors in unstructured data spaces using gradient-based latent interpolation

  title={Towards modelling hazard factors in unstructured data spaces using gradient-based latent interpolation},
  author={Tobias Weber and Michael Ingrisch and Bernd Bischl and David R{\"u}gamer},
The application of deep learning in survival analysis (SA) allows utilizing unstructured and high-dimensional data types uncommon in traditional survival methods. This allows to advance methods in fields such as digital health, predictive maintenance, and churn analysis, but often yields less interpretable and intuitively understandable models due to the black-box character of deep learning-based approaches. We close this gap by proposing 1) a multi-task variational autoencoder (VAE) with… 

Figures from this paper

DeepPAMM: Deep Piecewise Exponential Additive Mixed Models for Complex Hazard Structures in Survival Analysis
DeepPAMM is proposed, a versatile deep learning framework that is well-founded from a statistical point of view, yet with enough flexibility for modeling complex hazard structures and is competitive with other machine learning approaches with respect to predictive performance while maintaining interpretability.


A scalable discrete-time survival model for neural networks
This paper describes a discrete-time survival model designed to be used with neural networks, which it refers to as Nnet-survival, which is implemented in the Keras deep learning framework and source code for the model and several examples is available online.
Dynamic-DeepHit: A Deep Learning Approach for Dynamic Survival Analysis With Competing Risks Based on Longitudinal Data
A novel deep learning approach that is able to successfully address current limitations of standard statistical approaches such as landmarking and joint modeling is developed, and shows that Dynamic-DeepHit provides a drastic improvement in discriminating individual risks of different forms of failures due to cystic fibrosis.
DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network
The predictive and modeling capabilities of DeepSurv will enable medical researchers to use deep neural networks as a tool in their exploration, understanding, and prediction of the effects of a patient’s characteristics on their risk of failure.
DeepHit: A Deep Learning Approach to Survival Analysis With Competing Risks
A very different approach to survival analysis, DeepHit, that uses a deep neural network to learn the distribution of survival times directly and achieves large and statistically significant performance improvements over previous state-of-the-art methods.
Deep learning cardiac motion analysis for human survival prediction
A fully convolutional neural network is used to create time-resolved three-dimensional dense segmentations of heart images that can efficiently predict human survival.
Image-based Survival Analysis for Lung Cancer Patients using CNNs.
This work shows that by simplifying survival analysis to median survival classification, convolutional neural networks can be trained with small batch sizes and learn features that predict survival equally well as end-to-end hazard prediction networks.
Auto-Encoding Variational Bayes
A stochastic variational inference and learning algorithm that scales to large datasets and, under some mild differentiability conditions, even works in the intractable case is introduced.
Autoencoding beyond pixels using a learned similarity metric
An autoencoder that leverages learned representations to better measure similarities in data space is presented and it is shown that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic.
Deep Recurrent Survival Analysis
A Deep Recurrent Survival Analysis model is proposed which combines deep learning for conditional probability prediction at fine-grained level of the data, and survival analysis for tackling the censorship, and shows great advantages over the previous works on fitting various sophisticated data distributions.
Adversarial Latent Autoencoders
Autoencoder networks are unsupervised approaches aiming at combining generative and representational properties by learning simultaneously an encoder-generator map. Although studied extensively, the