Improving Marvel Hero Classification through Dataset Curation
DOI:
https://doi.org/10.61173/3gzt6303Keywords:
Marvel superheroes, computer vision, dataset curation, neural network configuration, Edge Impulse, image recognition, Space Shuttle Challenger disaster, O-ring failure, Bayesian logistic regression, temperature dependency, probabilistic modeling, aerospace engineering, safety assessmentAbstract
lores the development of a computer vision model for classifying Marvel superheroes such as Black Widow, Hulk, Iron Man, and Spider-Man. Utilizing a curated dataset sourced from Kaggle, the research emphasizes the critical role of dataset quality in refining model accuracy. Insights gained include adjustments to neural network configurations and leveraging Edge Impulse for enhanced performance. The findings highlight effective strategies for optimizing classification accuracy in complex image recognition tasks.PART 2:This part explores the application of Bayesian logistic regression to model the relationship between temperature and the probability of O-ring failure. By leveraging Bayesian inference techniques, analyzing historical data to quantify the risk associated with temperature variations and emphasize the importance of probabilistic approaches in safety-critical decision-making.The Space Shuttle Challenger disaster on January 28, 1986, remains a poignant case study in aerospace engineering failure. The investigation concluded that the failure of O-ring seals in cold temperatures led to the tragic loss of the shuttle and its crew.
References
suitable probabilistic model to capture the relationship between temperature and the probability of O-ring failure is a logistic regression model. Kaggle dataset: [https://www.kaggle.com/datasets/ Logistic Regression Model: hchen13/marvel-heroes] Model Formulation: Let ( T ) represent the temperature (in Fahrenheit) and ( F ) represent the binary outcome variable indicating O-ring failure (1 for failure, 0 for success). We assume that the probability of O-ring failure, ( p ), follows a logistic function: [ p(T) = frac{1}{1 + e^{-z}} ] where ( z ) is a linear combination of temperature and model parameters: [ z = beta_0 + beta_1 cdot T ] Here, ( beta_0 ) is the intercept parameter, and ( beta_1 ) is the coefficient associated with temperature. Bayesian Formulation: In a Bayesian framework, we assign prior distributions to the model parameters ( beta_0 ) and ( beta_1 ) and use Bayes’ theorem to update these priors based on observed data, yielding posterior distributions for the parameters. Priors: ( beta_0 sim Normal(mu_0, sigma_0^2) ) ( beta_1 sim Normal(mu_1, sigma_1^2) ) Likelihood: ( F_i sim Bernoulli(p(T_i)) ) for each observation ( i ) Posterior: ( beta_0, beta_1 | {T_i, F_i} sim ? ) (to be determined using MCMC methods) Implementation in PyMC: The logistic regression model can be implemented in PyMC using the pm.glm.GLM class, which provides a convenient interface for specifying generalized linear
Feynman, R. P. (1986). Report of the Presidential Commission on the Space Shuttle Challenger Accident. Washington, DC: Government Printing Office. Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2013). Bayesian Data Analysis (3rd ed.). Boca Raton, FL: Chapman and Hall/CRC. Bayesian formulation used in Bayesian logistic regression.
Downloads
Published
Issue
Section
License
Copyright (c) 2024 by the authors.
This work is licensed under a Creative Commons Attribution 4.0 International License.
