Elizabeth Silver — Causality and Causal Discovery
From the Melbourne Machine Learning and AI Meetup in September 2018 http://mlai.melbourne/
Talk starts at 11:35
Slides: http://www.lizziesilver.com/MLAI_September_2018.pdf
Elizabeth Silver — Causality and Causal Discovery
A tutorial & overview of methods for learning causal graphical models. We need causal models in several situations: to figure out which variables are confounders in observational studies; to suggest new experiments; and to predict the results of interventions if experiments are infeasible or unethical. I'll show you how to learn a causal model from observational data, and I'll show you the strengths and weaknesses of the methods.
Graphical models have a structure and a parameterisation. The structure represents the qualitative causal relationships - "what causes what". The parameterisation represents the strength and functional form of those relationships. You need to know the structure before you can learn the parameterisation.
I'll cover various algorithms for learning the model structure from observational data:
1. The PC algorithm ("Peter and Clark"), and PC-Stable
2. GES (Greedy Equivalence Search)
3. FCI (Fast Causal Inference)
4. LiNGAM (Linear Non-Gaussian Acyclic Models)
5. If we have time, more fancy stuff: latent variable models, non-linear causal additive models, SAT-solver methods, etc.
And I'll describe some of the major difficulties with causal inference:
* Validation
* Consistency
* Feature definition
* Measurement error
Speaker Bio:
Lizzie Silver is a Research Fellow at the University of Melbourne, where she analyses data for The SWARM Project, an effort to improve reasoning in human teams. She did her PhD at Carnegie Mellon University, on multitask methods for learning causal graphical models.
Meetup Event Page: https://www.meetup.com/Machine-Learning-AI-Meetup/events/xgxjvpyxmbxb/
Видео Elizabeth Silver — Causality and Causal Discovery канала Machine Learning and AI Meetup
Talk starts at 11:35
Slides: http://www.lizziesilver.com/MLAI_September_2018.pdf
Elizabeth Silver — Causality and Causal Discovery
A tutorial & overview of methods for learning causal graphical models. We need causal models in several situations: to figure out which variables are confounders in observational studies; to suggest new experiments; and to predict the results of interventions if experiments are infeasible or unethical. I'll show you how to learn a causal model from observational data, and I'll show you the strengths and weaknesses of the methods.
Graphical models have a structure and a parameterisation. The structure represents the qualitative causal relationships - "what causes what". The parameterisation represents the strength and functional form of those relationships. You need to know the structure before you can learn the parameterisation.
I'll cover various algorithms for learning the model structure from observational data:
1. The PC algorithm ("Peter and Clark"), and PC-Stable
2. GES (Greedy Equivalence Search)
3. FCI (Fast Causal Inference)
4. LiNGAM (Linear Non-Gaussian Acyclic Models)
5. If we have time, more fancy stuff: latent variable models, non-linear causal additive models, SAT-solver methods, etc.
And I'll describe some of the major difficulties with causal inference:
* Validation
* Consistency
* Feature definition
* Measurement error
Speaker Bio:
Lizzie Silver is a Research Fellow at the University of Melbourne, where she analyses data for The SWARM Project, an effort to improve reasoning in human teams. She did her PhD at Carnegie Mellon University, on multitask methods for learning causal graphical models.
Meetup Event Page: https://www.meetup.com/Machine-Learning-AI-Meetup/events/xgxjvpyxmbxb/
Видео Elizabeth Silver — Causality and Causal Discovery канала Machine Learning and AI Meetup
Показать
Комментарии отсутствуют
Информация о видео
19 октября 2018 г. 5:41:40
01:17:58
Другие видео канала
Causal Inference in Data Science From Prediction to Causation by Amit Sharma | DataEngConf NYC '16Stanford Seminar - Theories of inference for visual analysisIs it time for Automated Causal Discovery?Causal Discovery | Shawhin TalebiDiscovering contemporaneous and lagged causal relations in autocorrelated nonlinear time seriesKun Zhang, CMU: Causality, Independence, and Adaptive Prediction14. Causal Discovery With Linear Non-Gaussian Models Under Measurement ErrorBuilding and Using DAGs for Causal InferenceFixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence-Covered by Adel FodaCausal Inference — SUSAN ATHEY14. Causal Inference, Part 1A Framework for Understanding Unintended Consequences of Machine Learning — Laura SummersSelection Bias Graphically - Causal InferenceNonlinear Independent Component Analysis - Aapo HyvärinenSean Carroll on Causality and the Arrow of TimeDouble Machine Learning for Causal and Treatment EffectsMendelian randomisation for causal inference in epidemiologyKeynote: The Mathematics of Causal Inference: with Reflections on Machine LearningHindsight Experience Replay — Covered by Kristian HolsheimerKeynote 1: High dimensional Causal Inference -- Peter Bühlman