Valentyn Melnychuk
YOU?
Author Swipe
View article: An Orthogonal Learner for Individualized Outcomes in Markov Decision Processes
An Orthogonal Learner for Individualized Outcomes in Markov Decision Processes Open
Predicting individualized potential outcomes in sequential decision-making is central for optimizing therapeutic decisions in personalized medicine (e.g., which dosing sequence to give to a cancer patient). However, predicting potential ou…
View article: Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation
Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation Open
The conditional average treatment effect (CATE) is widely used in personalized medicine to inform therapeutic decisions. However, state-of-the-art methods for CATE estimation (so-called meta-learners) often perform poorly in the presence o…
View article: GDR-learners: Orthogonal Learning of Generative Models for Potential Outcomes
GDR-learners: Orthogonal Learning of Generative Models for Potential Outcomes Open
Various deep generative models have been proposed to estimate potential outcomes distributions from observational data. However, none of them have the favorable theoretical property of general Neyman-orthogonality and, associated with it, …
View article: Differentially Private Learners for Heterogeneous Treatment Effects
Differentially Private Learners for Heterogeneous Treatment Effects Open
Patient data is widely used to estimate heterogeneous treatment effects and thus understand the effectiveness and safety of drugs. Yet, patient data includes highly sensitive information that must be kept private. In this work, we aim to e…
View article: Efficient and Sharp Off-Policy Learning under Unobserved Confounding
Efficient and Sharp Off-Policy Learning under Unobserved Confounding Open
We develop a novel method for personalized off-policy learning in scenarios with unobserved confounding. Thereby, we address a key limitation of standard policy learning: standard policy learning assumes unconfoundedness, meaning that no u…
View article: Orthogonal Representation Learning for Estimating Causal Quantities
Orthogonal Representation Learning for Estimating Causal Quantities Open
End-to-end representation learning has become a powerful tool for estimating causal quantities from high-dimensional observational data, but its efficiency remained unclear. Here, we face a central tension: End-to-end representation learni…
View article: Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner
Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner Open
Estimating causal quantities from observational data is crucial for understanding the safety and effectiveness of medical treatments. However, to make reliable inferences, medical practitioners require not only estimating averaged causal q…
View article: DiffPO: A causal diffusion model for learning distributions of potential outcomes
DiffPO: A causal diffusion model for learning distributions of potential outcomes Open
Predicting potential outcomes of interventions from observational data is crucial for decision-making in medicine, but the task is challenging due to the fundamental problem of causal inference. Existing methods are largely limited to poin…
View article: Conformal Prediction for Causal Effects of Continuous Treatments
Conformal Prediction for Causal Effects of Continuous Treatments Open
Uncertainty quantification of causal effects is crucial for safety-critical applications such as personalized medicine. A powerful approach for this is conformal prediction, which has several practical benefits due to model-agnostic finite…
View article: G-Transformer for Conditional Average Potential Outcome Estimation over Time
G-Transformer for Conditional Average Potential Outcome Estimation over Time Open
Estimating potential outcomes for treatments over time based on observational data is important for personalized decision-making in medicine. Yet, existing neural methods for this task either (1) do not perform proper adjustments for time-…
View article: Causal machine learning for predicting treatment outcomes
Causal machine learning for predicting treatment outcomes Open
View article: A Neural Framework for Generalized Causal Sensitivity Analysis
A Neural Framework for Generalized Causal Sensitivity Analysis Open
Unobserved confounding is common in many applications, making causal inference from observational data challenging. As a remedy, causal sensitivity analysis is an important tool to draw causal conclusions under unobserved confounding with …
View article: Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimation
Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimation Open
State-of-the-art methods for conditional average treatment effect (CATE) estimation make widespread use of representation learning. Here, the idea is to reduce the variance of the low-sample CATE estimation by a (potentially constrained) l…
View article: Consistent End-to-End Estimation for Counterfactual Fairness
Consistent End-to-End Estimation for Counterfactual Fairness Open
Fairness in predictions is of direct importance in practice due to legal, ethical, and societal reasons. This is often accomplished through counterfactual fairness, which ensures that the prediction for an individual is the same as that in…
View article: Bayesian Neural Controlled Differential Equations for Treatment Effect Estimation
Bayesian Neural Controlled Differential Equations for Treatment Effect Estimation Open
Treatment effect estimation in continuous time is crucial for personalized medicine. However, existing methods for this task are limited to point estimates of the potential outcomes, whereas uncertainty estimates have been ignored. Needles…
View article: Estimating Average Causal Effects from Patient Trajectories
Estimating Average Causal Effects from Patient Trajectories Open
In medical practice, treatments are selected based on the expected causal effects on patient outcomes. Here, the gold standard for estimating causal effects are randomized controlled trials; however, such trials are costly and sometimes ev…
View article: Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Model
Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Model Open
Counterfactual inference aims to answer retrospective "what if" questions and thus belongs to the most fine-grained type of inference in Pearl's causality ladder. Existing methods for counterfactual inference with continuous outcomes aim a…
View article: Reliable Off-Policy Learning for Dosage Combinations
Reliable Off-Policy Learning for Dosage Combinations Open
Decision-making in personalized medicine such as cancer therapy or critical care must often make choices for dosage combinations, i.e., multiple continuous treatments. Existing work for this task has modeled the effect of multiple treatmen…
View article: Sharp Bounds for Generalized Causal Sensitivity Analysis
Sharp Bounds for Generalized Causal Sensitivity Analysis Open
Causal inference from observational data is crucial for many disciplines such as medicine and economics. However, sharp bounds for causal effects under relaxations of the unconfoundedness assumption (causal sensitivity analysis) are subjec…
View article: Fair Off-Policy Learning from Observational Data
Fair Off-Policy Learning from Observational Data Open
Algorithmic decision-making in practice must be fair for legal, ethical, and societal reasons. To achieve this, prior research has contributed various approaches that ensure fairness in machine learning predictions, while comparatively lit…
View article: Normalizing Flows for Interventional Density Estimation
Normalizing Flows for Interventional Density Estimation Open
Existing machine learning methods for causal inference usually estimate quantities expressed via the mean of potential outcomes (e.g., average treatment effect). However, such quantities do not capture the full information about the distri…
View article: Causal Transformer for Estimating Counterfactual Outcomes
Causal Transformer for Estimating Counterfactual Outcomes Open
Estimating counterfactual outcomes over time from observational data is relevant for many applications (e.g., personalized medicine). Yet, state-of-the-art methods build upon simple long short-term memory (LSTM) networks, thus rendering in…
View article: Estimating average causal effects from patient trajectories
Estimating average causal effects from patient trajectories Open
In medical practice, treatments are selected based on the expected causal effects on patient outcomes. Here, the gold standard for estimating causal effects are randomized controlled trials; however, such trials are costly and sometimes ev…
View article: Matching the Clinical Reality: Accurate OCT-Based Diagnosis From Few\n Labels
Matching the Clinical Reality: Accurate OCT-Based Diagnosis From Few\n Labels Open
Unlabeled data is often abundant in the clinic, making machine learning\nmethods based on semi-supervised learning a good match for this setting.\nDespite this, they are currently receiving relatively little attention in\nmedical image ana…
View article: Matching the Clinical Reality: Accurate OCT-Based Diagnosis From Few Labels
Matching the Clinical Reality: Accurate OCT-Based Diagnosis From Few Labels Open
Unlabeled data is often abundant in the clinic, making machine learning methods based on semi-supervised learning a good match for this setting. Despite this, they are currently receiving relatively little attention in medical image analys…
View article: Unsupervised Anomaly Detection for X-Ray Images
Unsupervised Anomaly Detection for X-Ray Images Open
Obtaining labels for medical (image) data requires scarce and expensive experts. Moreover, due to ambiguous symptoms, single images rarely suffice to correctly diagnose a medical condition. Instead, it often requires to take additional bac…