paper-with-me

Papers

Encoding Causal Macrovariables

2021-11-29 · Benedikt Höltgen

In many scientific disciplines, coarse-grained causal models are used to explain and predict the dynamics of more fine-grained systems. Naturally, such models require appropriate macrovariables. Automated procedures to detect suitable variables would be useful to leverage increasingly available high-dimensional observational datasets. This work introduces a novel algorithmic approach that is inspired by a new characterisation of causal macrovariables as information bottlenecks between microstates. Its general form can be adapted to address individual needs of different scientific goals. After a further transformation step, the causal relationships between learned variables can be investigated through additive noise models. Experiments on both simulated data and on a real climate dataset are reported. In a synthetic dataset, the algorithm robustly detects the ground-truth variables and correctly infers the causal relationships between them. In a real climate dataset, the algorithm robustly detects two variables that correspond to the two known variations of the El Nino phenomenon.

📄 PDF Abstract BibTeX arXiv:2111.14724

Code (1)

benedikthoeltgen/causal-macro 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Towards Effective Theory of LLMs: A Representation Learning Approach

2026-05-10 · Muhammed Ustaomeroglu, Guannan Qu arxiv

We propose Representational Effective Theory (RET), a framework for describing large language model computation in terms of learned macrostates rather than microscopic details. RET learns these macrostates from hidden-st…

Representation Learning

Causal interpretation rules for encoding and decoding models in neuroimaging

2015-11-15 · Sebastian Weichwald, Timm Meyer, Ozan Özdenizci, Bernhard Schölkopf 외

Causal terminology is often introduced in the interpretation of encoding and decoding models trained on neuroimaging data. In this article, we investigate which causal statements are warranted and which ones are not supp…

EEGElectroencephalogram (EEG)

Causal Discovery for Explainable AI: A Dual-Encoding Approach

2026-01-29 · Henry Salgado, Meagan R. Kendall, Martine Ceberio arxiv

Understanding causal relationships among features is fundamental for explaining machine learning model decisions. However, traditional causal discovery methods face challenges with categorical variables due to numerical …

An Empirical Study on the Impact of Positional Encoding in Transformer-based Monaural Speech Enhancement

2024-01-18 · Qiquan Zhang, Meng Ge, Hongxu Zhu, Eliathamby Ambikairajah 외

Transformer architecture has enabled recent progress in speech enhancement. Since Transformers are position-agostic, positional encoding is the de facto standard component used to enable Transformers to distinguish the o…

POSPositionSpeech Enhancement

Causality-Induced Positional Encoding for Transformer-Based Representation Learning of Non-Sequential Features

2025-09-20 · Kaichen Xu, Yihang Du, Mianpeng Liu, Zimu Yu 외 arxiv

Positional encoding is essential for supplementing transformer with positional information of tokens. Existing positional encoding methods demand predefined token/feature order, rendering them unsuitable for real-world d…

Representation Learning