LaDDer: Latent Data Distribution Modelling with a Generative Prior

1 code implementation • 31 Aug 2020 • Shuyu Lin, Ronald Clark

In this paper, we show that the performance of a learnt generative model is closely related to the model's ability to accurately represent the inferred \textbf{latent data distribution}, i. e. its topology and structural properties.

Representation Learning

Anomaly Detection for Time Series Using VAE-LSTM Hybrid Model

1 code implementation • IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2020 • Shuyu Lin, Ronald Clark, Robert Birke, Sandro Schönborn, Niki Trigoni, Stephen Roberts

In this work, we propose a VAE-LSTM hybrid model as an unsupervised approach for anomaly detection in time series.

Anomaly Detection Time Series +1

Balancing Reconstruction Quality and Regularisation in ELBO for VAEs

no code implementations • 9 Sep 2019 • Shuyu Lin, Stephen Roberts, Niki Trigoni, Ronald Clark

A trade-off exists between reconstruction quality and the prior regularisation in the Evidence Lower Bound (ELBO) loss that Variational Autoencoder (VAE) models use for learning.

WiSE-ALE: Wide Sample Estimator for Aggregate Latent Embedding

no code implementations • ICLR Workshop DeepGenStruct 2019 • Shuyu Lin, Ronald Clark, Robert Birke, Niki Trigoni, Stephen Roberts

In this paper, we present a new generative model for learning latent embeddings.

WiSE-ALE: Wide Sample Estimator for Approximate Latent Embedding

no code implementations • 16 Feb 2019 • Shuyu Lin, Ronald Clark, Robert Birke, Niki Trigoni, Stephen Roberts

Variational Auto-encoders (VAEs) have been very successful as methods for forming compressed latent representations of complex, often high-dimensional, data.