Our key result is that without assumptions about task-specific intuitions, explanations may potentially improve human understanding of model decision boundary, but they cannot improve human understanding of task decision boundary or model error.
no code implementations • 4 Feb 2022 • Tomas Geffner, Javier Antoran, Adam Foster, Wenbo Gong, Chao Ma, Emre Kiciman, Amit Sharma, Angus Lamb, Martin Kukla, Nick Pawlowski, Miltiadis Allamanis, Cheng Zhang
Causal inference is essential for data-driven decision making across domains such as business engagement, medical treatment or policy making.
Therefore, simple plausibility judgments on the maximum explanatory power of omitted variables (in explaining treatment and outcome variation) are sufficient to place overall bounds on the size of the bias.
A trained neural network can be interpreted as a structural causal model (SCM) that provides the effect of changing input variables on the model's output.
We develop a methodology, called Sayer, that leverages implicit feedback to evaluate and train new system policies.
Through extensive evaluation on a synthetic dataset and image datasets like MNIST, Fashion-MNIST, and Chest X-rays, we show that a lower OOD generalization gap does not imply better robustness to MI attacks.
Estimation of causal effects involves crucial assumptions about the data-generating process, such as directionality of effect, presence of instrumental variables or mediators, and whether all relevant confounders are observed.
Recent approaches have allowed for the denoising of single noisy images without access to any training data aside from that very image.
no code implementations • 11 Jan 2021 • Alexander Lavin, Ciarán M. Gilligan-Lee, Alessya Visnjic, Siddha Ganju, Dava Newman, Atılım Güneş Baydin, Sujoy Ganguly, Danny Lange, Amit Sharma, Stephan Zheng, Eric P. Xing, Adam Gibson, James Parr, Chris Mattmann, Yarin Gal
The development and deployment of machine learning (ML) systems can be executed easily with modern tools, but the process is typically rushed and means-to-an-end.
For example, in multi-stage settings where decisions are made in multiple screening rounds, we use our framework to derive the minimal distributions required to design a fair algorithm.
For many kinds of interventions, such as a new advertisement, marketing intervention, or feature recommendation, it is important to target a specific subset of people for maximizing its benefits at minimum cost or potential harm.
In addition, by restricting the features that can be modified for generating counterfactual examples, we find that the top-k features from LIME or SHAP are often neither necessary nor sufficient explanations of a model's prediction.
In addition to efficient statistical estimators of a treatment's effect, successful application of causal inference requires specifying assumptions about the mechanisms underlying observed data and testing whether they are valid, and to what extent.
We present a method for generating counterfactuals by incorporating a structural causal model (SCM) in an improved variant of Adversarially Learned Inference (ALI), that generates counterfactuals in accordance with the causal relationships between attributes of an image.
In the domain generalization literature, a common objective is to learn representations independent of the domain after conditioning on the class label.
Ranked #1 on Domain Generalization on Rotated Fashion-MNIST
The primary obstacle to developing technologies for low-resource languages is the lack of usable data.
For explanations of ML models in critical domains such as healthcare and finance, counterfactual examples are useful for an end-user only to the extent that perturbation of feature inputs is feasible in the real world.
To quantify this bias, we propose a general notion of $\eta$-infra-marginality that can be used to evaluate the extent of this bias.
We show that many popular methods, including back-door methods can be considered as weighting or representation learning algorithms, and provide general error bounds for their causal estimates.
Post-hoc explanations of machine learning models are crucial for people to understand and act on algorithmic predictions.
Digital Adherence Technologies (DATs) are an increasingly popular method for verifying patient adherence to many medications.
We present a method for estimating causal effects in time series data when fine-grained information about the outcome of interest is available.
The proposed algorithms use a best first search technique and report the solutions using an implicit representation ordered by cost.