# Contrastive representations of high-dimensional, structured treatments

arXiv: https://arxiv.org/abs/2411.19245 · HTML: https://mlsanity.com/paper/2411.19245/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 25 references that are in the map.
- First posted on arXiv 28 Nov 2024.

## Builds on (most cited first, 17 of 17)

- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style](https://mlsanity.com/paper/2106.04619/) (2021)
- [Recommendations as Treatments: Debiasing Learning and Evaluation](https://mlsanity.com/paper/1602.05352/) (2016)
- [Self-Supervised Learning Disentangled Group Representation as Feature](https://mlsanity.com/paper/2110.15255/) (2021)
- [Causal Effects of Linguistic Properties](https://mlsanity.com/paper/2010.12919/) (2020)
- [Causal Effect Inference for Structured Treatments](https://mlsanity.com/paper/2106.01939/) (2021)
- [D'ARTAGNAN: Counterfactual Video Generation](https://mlsanity.com/paper/2206.01651/) (2022)
- [Estimating Categorical Counterfactuals via Deep Twin Networks](https://mlsanity.com/paper/2109.01904/) (2021)
- [GraphITE: Estimating Individual Effects of Graph-structured Treatments](https://mlsanity.com/paper/2009.14061/) (2020)
- [Estimating long-term causal effects from short-term experiments and long-term observational data with unobserved confounding](https://mlsanity.com/paper/2302.10625/) (2023)
- [Disentangling Causal Effects from Sets of Interventions in the Presence of Unobserved Confounders](https://mlsanity.com/paper/2210.05446/) (2022)
- [Semiparametric Causal Sufficient Dimension Reduction Of Multidimensional Treatments](https://mlsanity.com/paper/1710.06727/) (2017)
- [Non-parametric identifiability and sensitivity analysis of synthetic control models](https://mlsanity.com/paper/2301.07656/) (2023)
- [Spillover Detection for Donor Selection in Synthetic Control Models](https://mlsanity.com/paper/2406.11399/) (2024)
- [Did I do that? Blame as a means to identify controlled effects in reinforcement learning](https://mlsanity.com/paper/2106.00266/) (2021)
- [Compositionally Equivariant Representation Learning](https://mlsanity.com/paper/2306.07783/) (2023)

## Built on by (most cited first, 7 of 7)

- [Agentic Personalisation of Cross-Channel Marketing Experiences](https://mlsanity.com/paper/2506.16429/) (2025)
- [Local Interference: Removing Interference Bias in Semi-Parametric Causal Models](https://mlsanity.com/paper/2503.18756/) (2025)
- [Learning Treatment Representations for Downstream Instrumental Variable Regression](https://mlsanity.com/paper/2506.02200/) (2025)
- [Quantum oracles give an advantage for identifying classical counterfactuals](https://mlsanity.com/paper/2512.13692/) (2025)
- [Predicting Causal Effects from Natural Language Queries using Structured Representations](https://mlsanity.com/paper/2605.29631/) (2026)
- [Entangled by Design: Spurious Intra-Variable Signal Routing in Tabular In-Context Learners](https://mlsanity.com/paper/2607.25532/) (2026)
- [From Training to Deployment: Post-Hoc Causal Feature Identification via Sensitivity Ratios](https://mlsanity.com/paper/2607.25546/) (2026)

## Abstract (arXiv)

Estimating causal effects is vital for decision making. In standard causal effect estimation, treatments are usually binary- or continuous-valued. However, in many important real-world settings, treatments can be structured, high-dimensional objects, such as text, video, or audio. This provides a challenge to traditional causal effect estimation. While leveraging the shared structure across different treatments can help generalize to unseen treatments at test time, we show in this paper that using such structure blindly can lead to biased causal effect estimation. We address this challenge by devising a novel contrastive approach to learn a representation of the high-dimensional treatments, and prove that it identifies underlying causal factors and discards non-causally relevant factors. We prove that this treatment representation leads to unbiased estimates of the causal effect, and empirically validate and benchmark our results on synthetic and real-world datasets.