# Dynamic β-VAEs for quantifying biodiversity by clustering optically recorded insect signals

@article{Rydhmer2021DynamicF,
title={Dynamic $\beta$-VAEs for quantifying biodiversity by clustering optically recorded insect signals},
author={Klas Rydhmer and Raghavendra Selvan},
journal={ArXiv},
year={2021},
volume={abs/2102.05526}
}
• Published 10 February 2021
• Computer Science
• ArXiv
While insects are the largest and most diverse group of animals, constituting ca. 80% of all known species, they are difficult to study due to their small size and similarity between species. Conventional monitoring techniques depend on time consuming trapping methods and tedious microscope-based work by skilled experts in order to identify the caught insect specimen at species, or even family, level. Researchers and policy makers are in urgent need of a scalable monitoring tool in order to… Expand

#### References

SHOWING 1-10 OF 62 REFERENCES
{m
• Computer Science
• ACML
• 2020
The master programme in Applied Geology aims to provide comprehensive knowledge based on various branches of Geology, with special focus on Applied geology subjects in the areas of Geomorphology, Structural geology, Hydrogeology, Petroleum Geologists, Mining Geology), Remote Sensing and Environmental geology. Expand
Acoustic auto-encoders for biodiversity assessment
• Computer Science
• Ecol. Informatics
• 2021
Advances in automatic identification of flying insects using optical sensors and machine learning.
It is demonstrated that it is possible to classify insects in flight, making it possible to optimize the application of insecticides in space and time, which will enable a technological leap in precision agriculture. Expand
• 2020
Balancing Reconstruction Error and Kullback-Leibler Divergence in Variational Autoencoders
• Computer Science
• IEEE Access
• 2020
This article shows that learning can be replaced by a simple deterministic computation, expressing the balancing factor in terms of a running average of the reconstruction error over the last minibatches, and sensibly outperforms all previous VAE architectures with comparable parameter capacity. Expand
Bark beetles as lidar targets and prospects of photonic surveillance.
Lidar results from a Swedish forest with controlled bark beetle release event are presented, and the capability of lidar to simultaneously monitor both insects and a pheromone plume mixed with chemical smoke governing the dispersal of many insects is demonstrated, suggesting entomological lidar is a promising tool for monitoring bark beetles. Expand
ControlVAE: Controllable Variational Autoencoder
A new non-linear PI controller is designed, a variant of the proportional-integral-derivative (PID) control, to automatically tune the hyperparameter (weight) added in the VAE objective using the output KL-divergence as feedback during model training. Expand
DynamicVAE: Decoupling Reconstruction Error and Disentangled Representation Learning
• Computer Science
• 2020
Evaluation results on three benchmark datasets demonstrate that DynamicVAE significantly improves the reconstruction accuracy while achieving disentanglement comparable to the best of existing methods. Expand
Entomological Scheimpflug lidar for estimating unique insect classes in-situ field test from Ivory Coast
• Environmental Science
• 2020
Acquisition of entomological data with high-frequency lidar is an emerging research field in rapid development. The technique offers very high numbers of observations per time unit, suitable forExpand
Lidar reveals activity anomaly of malaria vectors during pan-African eclipse
In situ lidar classification of several insect families and their sexes based on their modulation signatures is demonstrated and how lidar-based monitoring of distinct mosquito activities could advance the understanding of vector ecology is demonstrated. Expand