Classifying the unknown: Discovering novel gravitational-wave detector glitches using similarity learning

Coughlin, S. B.; Bahaadini, Sara; Rohani, Neda; Zevin, M.; Patane, Oli; Harandi, Mahboobeh; Jackson, Corey; Noroozi, Vahid; Allen, Sarah; Areeda, J. S.; Coughlin, M. W.; Ruiz, Pablo; Berry, C. P. L.; Crowston, Kevin; Katsaggelos, Aggelos K.; Lundgren, A. P.; Østerlund, Carsten; Smith, J. R.; Trouille, Laura; Kalogera, V.

doi:10.1103/physrevd.99.082002

Cited by 52 publications

(38 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We anticipate that future research may extend our findings to study how population dynamics can improve both the stability and accuracy of category systems in domains with objective truth conditions. In particular, we anticipate that future studies may apply our findings to address challenging issues in content moderation and classification, for instance to eliminate individual biases in large-scale citizen science efforts and related human crowdsourcing tasks, such as Galaxy Zoo 43 or Gravity Spy 44 , and to improve consistency in the classification of acceptable and unacceptable content on social media 45 .…”

Section: Discussionmentioning

confidence: 99%

Experimental evidence for scale-induced category convergence across populations

2021

View full text Add to dashboard Cite

Individuals vary widely in how they categorize novel and ambiguous phenomena. This individual variation has led influential theories in cognitive and social science to suggest that communication in large social groups introduces path dependence in category formation, which is expected to lead separate populations toward divergent cultural trajectories. Yet, anthropological data indicates that large, independent societies consistently arrive at highly similar category systems across a range of topics. How is it possible for diverse populations, consisting of individuals with significant variation in how they categorize the world, to independently construct similar category systems? Here, we investigate this puzzle experimentally by creating an online “Grouping Game” in which we observe how people in small and large populations collaboratively construct category systems for a continuum of ambiguous stimuli. We find that solitary individuals and small groups produce highly divergent category systems; however, across independent trials with unique participants, large populations consistently converge on highly similar category systems. A formal model of critical mass dynamics in social networks accurately predicts this process of scale-induced category convergence. Our findings show how large communication networks can filter lexical diversity among individuals to produce replicable society-level patterns, yielding unexpected implications for cultural evolution.

show abstract

Section: Discussionmentioning

confidence: 99%

Experimental evidence for scale-induced category convergence across populations

2021

View full text Add to dashboard Cite

show abstract

“…Finally, there is also a growing body of work which uses CNNs for various tasks that are di erent from but related to a gravitational-wave search, such as glitch classi cation (e.g., [45][46][47][48][49]) or parameter estimation (e.g., [50]). Furthermore, Dreissigacker et al [51] recently presented a proofof-principle study on using convolutional neural networks to search for continuous gravitational waves.…”

Section: Existing Cnn-based Approachesmentioning

confidence: 99%

Convolutional neural networks: A magic bullet for gravitational-wave detection?

et al. 2019

View full text Add to dashboard Cite

In the last few years, machine learning techniques, in particular convolutional neural networks, have been investigated as a method to replace or complement traditional matched ltering techniques that are used to detect the gravitational-wave signature of merging black holes. However, to date, these methods have not yet been successfully applied to the analysis of long stretches of data recorded by the Advanced LIGO and Virgo gravitational-wave observatories. In this work, we critically examine the use of convolutional neural networks as a tool to search for merging black holes. We identify the strengths and limitations of this approach, highlight some common pitfalls in translating between machine learning and gravitational-wave astronomy, and discuss the interdisciplinary challenges. In particular, we explain in detail why convolutional neural networks alone cannot be used to claim a statistically signi cant gravitational-wave detection. However, we demonstrate how they can still be used to rapidly ag the times of potential signals in the data for a more detailed follow-up. Our convolutional neural network architecture as well as the proposed performance metrics are better suited for this task than a standard binary classi cations scheme. A detailed evaluation of our approach on Advanced LIGO data demonstrates the potential of such systems as trigger generators. Finally, we sound a note of caution by constructing adversarial examples, which showcase interesting "failure modes" of our model, where inputs with no visible resemblance to real gravitational-wave signals are identi ed as such by the network with high con dence.

show abstract

“…In recent years, the amount and complexity of data from large detectors such as the Laser Interferometer Gravitational-Wave Observatory (LIGO) have grown enormously and exceeded the time capacities of volunteers. In a search for a new computational technology, researchers who established the Gravity Spy project tested a joint workflow between citizen scientists who identified novel glitches and machine learning techniques (Coughlin et al 2019 ; cf. also Franzen et al, Chap.…”

Section: Research Approachesmentioning

confidence: 99%

Citizen Science in the Natural Sciences

Frigerio

Richter

Per

et al. 2021

The Science of Citizen Science

View full text Add to dashboard Cite

The natural sciences include the life and physical sciences and study nature through observing and understanding phenomena, testing hypotheses, and performing experiments. Key principles such as reliability, validity, objectivity, and predictability are achieved through transparent assumptions, methods, data, and interpretations as well as multidisciplinarity.In this chapter we present insights into the genesis of citizen science in the natural sciences and reflect on the intellectual history of the natural sciences in relation to citizen science today. Further, we consider the current scientific approaches and achievements of natural science projects, which are applying citizen science to address empirical and/or theoretical research, focusing on monitoring programmes. Presenting examples and case studies, we focus on the key characteristics of the scientific inquiries being investigated in the natural sciences through citizen science. Finally, we discuss the consequences of engagement in scientific processes in relation to the future of natural scientists in a complex world.

show abstract

Classifying the unknown: Discovering novel gravitational-wave detector glitches using similarity learning

Cited by 52 publications

References 42 publications

Experimental evidence for scale-induced category convergence across populations

Experimental evidence for scale-induced category convergence across populations

Convolutional neural networks: A magic bullet for gravitational-wave detection?

Citizen Science in the Natural Sciences

Contact Info

Product

Resources

About