Express improves causal attention guarantees for language models.
problem Improving causal attention guarantees for language models.
method Introducing Express, a tool for converting non-causal attention into causal with matching guarantees.
result Express improves causal attention guarantees to log^(3/2)(n)/s with minimal memory and compression overhead.
Paper estimates non-causal graphical models using covariance extension and transportation distance.
problem Estimating non-causal graphical models with smoothing relations.
method Proposes a covariance extension problem and uses transportation distance to minimize error with white noise.
result Solution is a double-sided autoregressive non-causal graphical model.
Casper uses causal graph neural networks to improve spatiotemporal time series imputation.
problem Imputing missing values in spatiotemporal time series with confounders and non-causal correlations.
method Casper introduces a novel Prompt Based Decoder (PBD) and Spatiotemporal Causal Attention (SCA) to block confounders and discover causal relationships.
result Casper outperforms baselines and effectively discovers causal relationships in spatiotemporal time series imputation.
Paper develops a new inequality for non-causal machine learning.
problem Current concentration inequalities cannot be applied to non-causal machine learning.
method Develops a framework for non-causal random fields and proves a Hoeffding-type inequality.
result Obtains a Hoeffding-type concentration inequality for non-causal random fields.
OSA overcomes instability in skipless Transformers.
problem Instability in skipless Transformers using Softmax Self-Attention.
method OSA parametrizes attention matrix to be orthogonal via skew-symmetric matrix exponential.
result OSA allows for training non-causal Transformers without skip connections and normalisation layers.
In this paper, we aim to develop a unified view of causal and non-causal feature selection methods. The unified view will fill in the gap in the research of the relation between the two types of methods. Based on the Bayesian network framework and information theory, we first show that causal and non-causal feature sel…
This paper introduces a new task to better understand Transformers in quantitative contexts.
problem Understanding Transformers in high-stakes quantitative and scientific applications.
method Introduces a novel contextual counting task and analyzes it with causal and non-causal Transformer architectures.
result Causal attention is better suited for the contextual counting task, and no positional embeddings lead to the best accuracy.
New method tests Granger non-causality in panel data with cross-sectional dependencies.
problem Testing Granger non-causality in panel data with cross-sectional dependencies.
method Proposes a new approach to aggregate p-values from panel members to test Granger non-causality, showing lower FDR.
result Our approach discovers true causal relations in panel data, unlike state-of-the-art methods.
The paper proposes a method to stabilize predictions by identifying causal variables using a seed variable.
problem Stable prediction across unknown test data with potential spurious correlations.
method Conditional independence test based algorithm using a seed variable to separate causal from non-causal variables.
result The algorithm precisely separates causal and non-causal variables for stable prediction across test data.
New study finds best language model architecture and pretraining objective for zero-shot tasks.
problem Evaluating which language model architectures and pretraining objectives best enable zero-shot generalization.
method Compared three model architectures and two pretraining objectives across 170 billion tokens, with and without finetuning.
result Causal decoder-only models trained on autoregressive language modeling exhibit strongest zero-shot generalization.
Fairness of classification and regression has received much attention recently and various, partially non-compatible, criteria have been proposed. The fairness criteria can be enforced for a given classifier or, alternatively, the data can be adapated to ensure that every classifier trained on the data will adhere to d…
New method allows real-time audio synthesis using non-causal convolutions.
problem Real-time audio synthesis limitations due to offline model constraints.
method Post-training reconfiguration of non-causal models for real-time buffer-based processing.
result Non-causal streaming models can be transformed from offline-trained models without quality loss.
New method reduces diffusion model function evaluations for discrete data.
problem High computational burden in generating samples from masked diffusion models.
method Modified causal attention mask and speculative sampling mechanism for non-factorized predictions.
result Achieved ~2x reduction in required network forward passes.
We consider linear models where d potential causes X1,...,Xd are correlated with one target quantity Y and propose a method to infer whether the association is causal or whether it is an artifact caused by overfitting or hidden common causes. We employ the idea that in the former case the vector of regression c…
New methods calibrate causal estimates using standard predictive models.
problem Calibrating causal treatment effect estimates.
method Developed algorithms to transform causal estimation into standard calibration.
result General algorithms for causal calibration using standard predictive models.
Novel framework uses causality for financial forecasting.
problem Balancing invariance and prediction accuracy in financial time series.
method Causality-inspired models for forecasting asset returns.
result Efficacy in stable and accurate predictions, especially in turbulent markets.
Predictive models can be used for causal inference with feature selection.
problem Limitations of predictive models in interpreting causal relationships.
method Constrained learning process by selecting features according to Pearl's backdoor adjustment criterion.
result Causal models provide near unbiased effect estimates and better generalization.
Proposes Causal Loss to improve machine learning models' causal inference.
problem Machine learning algorithms often fail to capture causal relationships when data is inconsistent.
method Introduces Causal Loss, a model-agnostic loss function that enhances interventional capabilities.
result Causal Loss improves non-causal associative models to have interventional capabilities.
New clustering method considers causal fairness to avoid bias.
problem Clustering algorithms can unintentionally propagate unfair disparities.
method Integrates causal fairness metrics into clustering algorithms.
result Demonstrates efficacy on datasets with known unfair biases.
Enhances causal estimation using unlabeled offline ICU data.
problem Assessing unmeasured physiological variables in new ICU patients.
method Three-stage approach: non-causal and causal estimators, causal filter, and prediction for new patients.
result Enhanced causal estimation for new ICU patients using offline data.
Summary statistics of genome-wide association studies (GWAS) teach causal relationship between millions of genetic markers and tens and thousands of phenotypes. However, underlying biological mechanisms are yet to be elucidated. We can achieve necessary interpretation of GWAS in a causal mediation framework, looking to…
New model improves data augmentation for causal tasks.
problem Optimizing causal models robustly under Wasserstein distances.
method Proposes a new G-Causal Normalizing Flow architecture.
result Empirically outperforms standard generative models.
Behavioral cloning reduces policy learning to supervised learning by training a discriminative model to predict expert actions given observations. Such discriminative models are non-causal: the training procedure is unaware of the causal structure of the interaction between the expert and the environment. We point out …
We present a Hawkes model approach to foreign exchange market in which the high frequency price dynamics is affected by a self exciting mechanism and an exogenous component, generated by the pre-announced arrival of macroeconomic news. By focusing on time windows around the news announcement, we find that the model is …
While most classical approaches to Granger causality detection repose upon linear time series assumptions, many interactions in neuroscience and economics applications are nonlinear. We develop an approach to nonlinear Granger causality detection using multilayer perceptrons where the input to the network is the past t…
Researchers use human-in-the-loop to create counterfactually augmented data, improving model performance.
problem Creating ML models less reliant on spurious patterns in NLP datasets.
method A human-in-the-loop process to curate counterfactually augmented data (CAD), prohibiting unnecessary edits.
result Models trained on CAD appear to rely less on semantically irrelevant words and generalize better out of domain.
Real-time speech enhancement model removes various noises and reverb.
problem Real-time speech enhancement in noisy environments.
method Causal speech enhancement model using encoder-decoder architecture with skip-connections, optimized in time and frequency domains.
result The model matches state-of-the-art performance while working directly on raw waveform.
Traditional pairwise sequence alignment is based on matching individual samples from two sequences, under time monotonicity constraints. However, in many application settings matching subsequences (segments) instead of individual samples may bring in additional robustness to noise or local non-causal perturbations. Thi…
New method learns unbiased treatment representations from structured high-dimensional data.
problem Estimating causal effects from high-dimensional, structured treatments.
method Contrastive learning approach to learn unbiased treatment representations.
result The method identifies causal factors and discards non-causal ones, leading to unbiased causal effect estimates.
Unobserved confounding is a major hurdle for causal inference from observational data. Confounders---the variables that affect both the causes and the outcome---induce spurious non-causal correlations between the two. Wang & Blei (2018) lower this hurdle with "the blessings of multiple causes," where the correlation st…
Dynamic Structural Causal Models handle time-dependent systems with cycles and latent confounding.
problem Representing and analyzing systems of Stochastic Differential Equations (SDEs) with DSCMs.
method Define time-splitting and subsampling operations to analyze DSCMs of SDEs, and apply existing causal discovery algorithms to time-series data.
result DSCMs provide a graphical Markov property for SDEs and enable identification of time-dependent causal effects.
Develops a new method to discover causal relationships from nonstationary time series data.
problem Challenges in inferring causal relationships from observational data, especially for nonstationary time series.
method State-Dependent Causal Inference (SDCI) for conditionally stationary time series.
result SDCI can recover underlying causal dependencies with provable identifiability for state-dependent causal structures.
Method constructs prediction intervals for time-varying individual treatment effects.
problem Accurately quantify uncertainty of individual treatment effects across multiple decision points.
method Conformal inference techniques for time-varying ITEs with weaker assumptions.
result Guaranteed lower bound for coverage dependent on data non-exchangeability.
Recourse explanations can become invalid if collective actions change statistical data.
problem Recourse explanations may become invalid due to collective behavior changing data statistics.
method Formal characterization of conditions under which recourse explanations remain valid under performativity.
result Recourse actions may become invalid if they are influenced by or intervene on non-causal variables.
MO-CBO optimizes multiple outcomes in causal systems with minimal data.
problem Optimizing multiple outcomes in causal systems with limited data.
method Decomposes MO-CBO into multi-objective optimization tasks and uses relative hypervolume improvement for sequential intervention balancing.
result MO-CBO outperforms traditional multi-objective Bayesian optimization in causal settings.
CGRL improves graph neural networks' OOD generalization by blocking spurious correlations.
problem Graph Neural Networks struggle with out-of-distribution data due to learning spurious correlations.
method Formulates a causal graph, uses backdoor adjustment, and introduces a loss replacement strategy.
result Significantly improves OOD generalization of GNNs, stabilizing mutual information learning.
CausAdv detects adversarial examples using causal reasoning.
problem Vulnerability of CNNs to adversarial perturbations.
method Causal framework based on counterfactual reasoning.
result Adversarial examples exhibit different CI distributions compared to clean samples.
COTA learns abstraction maps from data without complete SCM knowledge.
problem Learning causally consistent representations at different resolutions.
method Multi-marginal Optimal Transport (OT) with do-calculus constraints and interventional cost.
result COTA outperforms non-causal and independent formulations on synthetic and real-world problems.
New interpretation of attention in Transformers and Graph Attention Networks.
problem Understanding and improving attention mechanisms in deep learning models.
method Decomposed attention into a kernel and a normalization term; generalized the kernel function and norm.
result Generalized attention leads to better performance on various tasks.
New approach improves multi-head attention by making heads less similar.
problem Multi-head attention can lead to similar features, reducing model expressiveness.
method Proposes a non-parametric approach using Bayesian techniques to make heads repel each other.
result Improves feature diversity, leading to better representations and performance.
Develops a framework for distributional Granger causality
problem Identifying predictive dependence in time series beyond Gaussian settings
method Using finite collection of channel-specific restrictions
result Identifies distributional Granger non-causality through testable hypotheses
Regularizes attention scores in vision transformers using bootstrapping.
problem Noisy and diffused attention maps in ViT limit interpretability.
method Statistical learning techniques, bootstrapping of attention scores.
result Improves shrinkage and sparsity of attention scores.
Random forests with attention and self-attention improve regression performance.
problem Improving regression model performance on various datasets.
method Proposes new models using attention and self-attention mechanisms to solve regression problems.
result The models improve model performance on many datasets.
Aligns attention distributions for improved accuracy and robustness.
problem Improving the accuracy and robustness of neural networks using attention mechanisms.
method Alignment attention that encourages key and query distributions to match within each head.
result Alignment attention leads to better accuracy, uncertainty estimation, and robustness across various tasks.
A framework for transformer attention layers derived from SVR.
problem Developing principled attention mechanisms for transformers.
method Mapping self-attention to SVR, deriving new attention types.
result Improved transformer performance and efficiency.
Elliptical Attention improves transformer performance by focusing on contextually relevant features.
problem Transformer models suffer from representation collapse and are vulnerable to contaminated samples.
method Uses Mahalanobis distance to define hyper-ellipsoidal neighborhoods for attention weights.
result Elliptical Attention reduces representation collapse and enhances model robustness.
Kernel PCA explains self-attention mechanisms in deep learning models.
problem Understanding and explaining self-attention mechanisms in deep learning models.
method Deriving self-attention from kernel principal component analysis (kernel PCA).
result RPC-Attention, a robust attention mechanism, outperforms softmax attention in various tasks.
Study examines asset pricing using various attention models, finding global self-attention and sliding window sparse attention models perform well.
problem Traditional asset pricing models miss temporal dependency and short memory issues.
method Investigates RNN attention models with various attention mechanisms for large-cap US stocks.
result Global self-attention and sliding window sparse attention models outperform in deriving returns and hedging risks, especially during the pandemic.