Data to the Rescue! Predicting and Preventing Accidents at Sea
Offered By: MLCon | Machine Learning Conference via YouTube
Course Description
Overview
Explore the cutting-edge application of machine learning in maritime safety through this conference talk from ML Conference 2019. Discover how data science and advanced algorithms are revolutionizing the marine insurance industry, predicting and potentially preventing accidents at sea. Learn about the integration of ship behavior data, including location, speed, maps, and weather information, into sophisticated machine learning models. Gain insights into the challenges of introducing modern technology to a traditional industry and the importance of model interpretability using SHAP. Delve into topics such as evolutionary computation, operational profiles, anomaly detection, and time series analysis in the context of maritime risk assessment. Understand how these innovations can significantly reduce the human, financial, and environmental costs associated with maritime accidents.
Syllabus
Intro
Evolutionary Computation
Introduction
Why Ships
Accidents at Sea
Risky Drivers
Operational Profiles
Data
Windward
Context
What a ship is doing
Geography
Draft
Night
Maps
Movie
Separation Zone
Anomaly Detection
Weather
Operational Profile
Accidents Database
Machine Learning Models
Deep Learning
Time Series Data
Sentence Completion
Time Series
Network
Predicting Ship Class
Predicting Kerch Accident
Insurance Companies
Lloyds
Wayne Ward
How do we gain trust
How well we become impactful
How we do it
Probability vs Loss
Trust
Use
SHAP
Conclusion
Questions
Taught by
MLCon | Machine Learning Conference
Related Courses
Machine Learning Interpretable: SHAP, PDP y permutacionCoursera Project Network via Coursera Machine Learning and AI Foundations: Producing Explainable AI (XAI) and Interpretable Machine Learning Solutions
LinkedIn Learning Guided Project: Predict World Cup Soccer Results with ML
IBM via edX What is Interpretable Machine Learning - ML Explainability - with Python LIME Shap Tutorial
1littlecoder via YouTube How Can I Explain This to You? An Empirical Study of Deep Neural Net Explanation Methods - Spring 2021
University of Central Florida via YouTube