Evolution of Novel Activation Functions in Neural Network Training with Applications to Classification of Exoplanets

Abstract

We present analytical exploration of novel activation functions as consequence of integration of several ideas leading to implementation and subsequent use in habitability classification of exoplanets. Neural networks, although a powerful engine in supervised methods, often require expensive tuning efforts for optimized performance. Habitability classes are hard to discriminate, especially when attributes used as hard markers of separation are removed from the data set. The solution is approached from the point of investigating analytical properties of the proposed activation functions. The theory of ordinary differential equations and fixed point are exploited to justify the "lack of tuning efforts" to achieve optimal performance compared to traditional activation functions. Additionally, the relationship between the proposed activation functions and the more popular ones is established through extensive analytical and empirical evidence. Finally, the activation functions have been implemented in plain vanilla feed-forward neural network to classify exoplanets.

41 pages, 11 figures

References in corpus (20)

• Scikit-learn: Machine Learning in Python
• SMOTE: Synthetic Minority Over-sampling Technique
• Planetary Candidates Observed by Kepler, III: Analysis of the First 16 Months of Data
• Prevalence of Earth-size planets orbiting Sun-like stars
• One or more bound planets per Milky Way star from microlensing observations
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• The cometary composition of a protoplanetary disk as revealed by complex cyanides
• Lyman-alpha Emission from a Luminous z=8.68 Galaxy: Implications for Galaxies as Tracers of Cosmic Reionization
• Exploring Exoplanet Populations with NASA's Kepler Mission
• The Galactic Habitable Zone I. Galactic Chemical Evolution
• Disequilibrium biosignatures over Earth history and implications for detecting exoplanet life
• Characterizing the Cool KOIs IV: Kepler-32 as a prototype for the formation of compact planetary systems throughout the Galaxy
• Superhabitable Worlds
• GJ 832c: A super-earth in the habitable zone
• Nomads of the Galaxy
• The language of exoplanet ranking metrics needs to change
• Planets in the Early Universe
• Kepler's Earth-like Planets Should Not Be Confirmed Without Independent Detection: The Case of Kepler-452b
• Age Aspects of Habitability
• A Comparative Analysis of the Cobb-Douglas Habitability Score (CDHS) with the Earth Similarity Index (ESI)

Cited by in corpus (2)

• Postulating Exoplanetary Habitability via a Novel Anomaly Detection Method
• Quantifying the Classification of Exoplanets: in Search for the Right Habitability Metric