Estimating the effect of ‘diverse’ team compositions on Dota 2 game outcomes using Inverse Probability Weighting

Commonly, when researchers are figuring out the effect of a putative cause, additional variables influence the cause and the effect. These are called confounders, and they obfuscate causal relationships. Inverse Probability Weighting is a method that can be applied to remove confounding and show a causal effect. This study aims to determine if...

Understanding Risk Extrapolation (REx) and when it finds Invariant Relationships

Generalizing models for new unknown datasets is a common problem in machine learning. Algorithms that perform well for test instances with the same distribution as their training dataset often perform severely on new datasets with a different distribution. This problem is caused by distributional shifts between the training of the model and...

Group Distributionally Robust Optimization for Solving Out-Of-Domain Generalization and Finding Causal Invariant Relationships

Out-of-Domain (OOD) generalization is a challenging problem in machine learning about learning a model from one or more domains and making the model perform well on an unseen domain. Empirical Risk Minimization (ERM), the standard machine learning method, suffers from learning spurious correlation in the training domain, therefore may perform...

Causal Inference in DotA 2 when estimated through randomized data

Strategy games could be considered as an amazing playground for using Causal inference methods. The complex nature of the data and the built-in randomization help with testing causal inference in a scenario where in reality it would be hard and expensive. Randomized data in coherence with causal inference is well documented and tested, but not...

Validating the win-rate of heroes in Dota 2 using instrumental variable estimation

This research provides additional insights in to when instrumental variable estimation is a proper method to use when investigating or removing causal effects in randomized experiments. this is done by using instrumental variables on the game Dota 2, in which win-rates of a couple heroes are validated and reviewed to test the effectiveness and...

Evaluating the Performance of the Model Selection with Average ECE and Naive Calibration in Out-of-Domain Generalization Problems for Binary Classifiers

Out-of-domain (OOD) generalization refers to learning a model from one or more different but related domain(s) that can be used in an unknown test domain. It is challenging for existing machine learning models. Several methods have been proposed to solve this problem, and multi-domain calibration is one of these methods. Model selection with the...

Finding the causal effect of picking Slark in the game of Dota 2 on the final match outcome based on in-match events using the front-door adjustment

The front-door adjustment is a causal inference method with which it is possible to determine the causal effect of applying a treatment given a setting which satisfies the front-door criterion. This involves having a mediator through which all the causal effect flows from treatment to outcome. The front-door adjustment adjusts for confounders...

Discovering the effect of hero choice on the outcome of a Dota 2 game

Dota 2 is one of the most popular MOBA (Multiplayer Online Battle Arena) games being played today. A Dota 2 match is played by two teams of 5 players. The main goal of the game is to destroy the opposing team’s Ancient tower, the team that manages to do so, wins the game. An essential part of a match is the hero selection phase before it starts....

Can Invariant Risk Minimization resist the temptation of learning spurious correlations?

Learning algorithms can perform poorly in unseen environments when they learn<br/>spurious correlations. This is known as the out-of-domain (OOD) generalization problem. Invariant Risk Minimization (IRM) is a method that attempts to solve this problem by learning invariant relationships. Motivating examples as well as counterexamples have been...

An empirical study of the effects of unconfoundedness on the performance of Propensity Score Matching

The purpose of this research is to analyze the performance of Propensity Score Matching, a causal inference method for causal effect estimation. More specifically, investigate how Propensity Score Matching reacts to breaking the unconfoundedness assumption, one of its core conceptual pillars. This has been achieved by running PSM on synthetic...

Honesty in Causal Forests, is it worth it ?

Causal machine learning is a relatively new field which tries to find a causal relation between the treatment and the outcome, rather than a correlation between the features and the outcome. To achieve this, many different models were proposed, one of which is the causal forest. Causal forest is made up of a random forest, with a different...

Treatment Effect Estimation of the DragonNet under Overlap Violations

The large amounts of observational data available nowadays have sparked considerable interest in learning causal relations from such data using machine learning methods. One recent method for doing this, which provided promising results, is the DragonNet (Shi et al., 2019), which utilises neural networks in order to estimate average treatment...

Empirical study of GANITE’s robustness to hidden confounders

An empirical study is performed exploring the sensitivity to hidden confounders of GANITE, a method for Individualized Treatment Effect (ITE) estimation. Most real world datasets do not measure all confounders and thus it is important to know how crucial this is in order to obtain comparable predictions. This is explored through the removal of...

Empirical Evaluation of the Performance of CEVAE under Misspecification of the Latent Dimensionality

Causal machine learning deals with the inference of causal relationships between variables in observational datasets. <br/>For certain datasets, it is correct to assume a causal graph where information about unobserved confounders can only be obtained through noisy proxies, and CEVAE aims to address this case. <br/>The number of dimensions of...

The prediction of extubation failure after surgery in pediatric patients with congenital heart disease

Background and aims<br/>The timing of extubation is a difficult decision for the medical team on the PICU. With negative impact on patient outcome when extubating too late or too early. The aim of this study was to create machine learning models for extubation failure prediction after surgery in patients with congenital heart disease. The goal...

Active Learning by Discrepancy Minimization: A Comparison of Active Learning Methods Motivated by Generalization Bounds

In many settings in practice it is expensive to obtain labeled data while unlabeled data is abundant. This is problematic if one wants to train accurate (supervised) predictive models. The main idea behind active learning is that models can perform better with less labeled data, if the model may choose the data from which it learns. Active...