CLEAR learns causal graphs from attention in recommender systems to explain user behavior.
problem Understanding why specific recommendations are made in recommender systems.
method CLEAR learns session-specific causal graphs from attention in pre-trained neural recommenders, addressing latent confounders.
result CLEAR provides counterfactual explanations that are shorter and more effective than naive methods.
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.
CInA method uses attention to improve causal inference.
problem Challenges in causal inference, especially in complex tasks.
method CInA method utilizes self-supervised causal learning with multiple unlabeled datasets and transformer-type architecture.
result CInA effectively generalizes to out-of-distribution datasets and various real-world datasets.
Transformers learn causal structure through gradient descent on self-attention mechanisms.
problem Understanding how transformers learn causal structure during training.
method In-context learning task and simplified two-layer transformer model.
result Gradient descent on a simplified transformer learns to encode latent causal graphs.
ISAHP discovers instance-level causal structures in event sequences.
problem Discovering fine-grained causal relationships in asynchronous, interdependent event sequences.
method ISAHP, a novel deep learning framework using self-attention mechanism.
result ISAHP meets Granger causality requirements and discovers complex causal structures.
MOCA uses modular attention to estimate causal effects from complex data.
problem Estimating causal effects from observational data with complex, non-linear, and high-dimensional treatment and outcome mechanisms.
method MOCA is a transformer-based framework that separates treatment and outcome modeling through modular design and one-way attention mechanism, with cutting-feedback to prevent outcome influence on treatment representations.
result MOCA outperforms classical estimators and machine learning approaches across various simulated and real-world scenarios.
InGRA models for efficient Granger causality learning in multivariate time series.
problem Efficiently modeling Granger causality in large-scale multivariate time series data.
method Inductive GRanger causal modeling (InGRA) framework with prototypical Granger causal attention.
result InGRA detects common causal structures and infers Granger causal structures for new individuals.
CgNN uses network structure as IVs to estimate causal effects in networks.
problem Hidden confounders complicate causal effect estimation in network data.
method CgNN combines GNNs and attention mechanisms to leverage network structure as IVs.
result CgNN effectively mitigates hidden confounder bias and improves causal effect estimation.
Proposes causal modeling for intersectional fairness in rankings.
problem Fairness in rankings, especially intersectional fairness.
method Causal modeling approach for intersectional fairness, flexible ranking computation.
result Experimental evaluation shows the approach's effectiveness under different assumptions.
GPT learns a causal world model from token predictions, validated in game sequences.
problem Does GPT implicitly learn a causal world model from token predictions?
method Derived a causal interpretation of GPT's attention mechanism and proposed zero-shot causal structure learning.
result GPT can generate legal next moves with high confidence for sequences with encoded causal structures, but fails for illegal moves.
Transformer model handles causal inference with DAG integration.
problem Complex causal structures and adaptability across various scenarios.
method Integrates DAGs into transformer's attention mechanism.
result Surpasses existing methods in estimating causal effects.
CauScale efficiently discovers causal relationships in large graphs.
problem Efficiency bottlenecks in causal discovery for large graphs.
method Neural architecture with reduction unit and tied attention weights.
result Achieves 99.6% mAP on in-distribution data and 84.4% on out-of-distribution data.
CaT-GNN improves credit card fraud detection by integrating causal reasoning into GNNs.
problem Credit card fraud detection overlooks causal structure of transactions.
method CaT-GNN combines causal invariant learning and temporal graph neural networks.
result CaT-GNN outperforms existing methods on various datasets.
Introduces recency bias to improve time-series forecasting.
problem Lack of recency bias in standard Transformer attention for time-series data.
method Reweights attention scores with a smooth heavy-tailed decay to emphasize nearby observations.
result Recency-biased attention consistently improves sequential modeling and achieves competitive performance on time-series forecasting benchmarks.
New model learns causal world dynamics from state space models.
problem Lack of causal world models in neural world modeling.
method State Space Models (SSM) with attention mechanisms.
result SSM can learn causal models of environments with equivalent performance.
CSHT predicts financial returns from news using a novel transformer model on a sphere.
problem Financial forecasting from news and sentiment.
method Granger-causal hypergraph structure, Riemannian geometry, causally masked Transformer attention.
result CSHT outperforms baselines in return prediction, regime classification, and asset ranking.
The study explores how Transformers predict the next token in a sequence.
problem Understanding the mechanism behind Transformers' autoregressive learning ability.
method Exploring the approximation ability of Transformers for next-token prediction through specific instances and a causal kernel descent method.
result Transformer models can learn context-dependent functions f for next-token prediction based on past and current observations. A new method infers causal structures and generates data without DAGs.
problem Modeling causal relationships without DAGs.
method Fixed-point approach on causally ordered variables, amortized TO inference, transformer-based SCM learning.
result The model learns TOs and SCMs from data, outperforming baselines.
Collectivistic countries influence Bitcoin returns more than individualistic ones during the pandemic.
problem The impact of public attention to COVID-19 on Bitcoin returns.
method Rolling and recursive-evolving algorithms to account for timing and estimation bias.
result Collectivistic countries have stronger causal impacts on Bitcoin returns.
Transformers predict price movements from limit order books.
problem Predicting price movements from limit order books.
method Causal convolutional network with masked self-attention.
result Significantly outperforms existing architectures on FI-2010 dataset.
FoundCause: Causal Discovery with Latent Confounders from Observational Data
problem Causal discovery from observational data
method FoundCause, an amortized causal discovery model trained on synthetic data
result FoundCause outperforms classical and amortized methods on real-world datasets
Introduces Causal Energy Minimization to understand Transformer layers.
problem Empirical parameterization of Transformer blocks remains largely unexplored.
method Causal Energy Minimization framework that recasts Transformer layers as optimization steps on conditional energy functions.
result Identifies design space for Transformer layers including weight sharing and energy-based interpretations.
New framework learns disentangled causal representations from observed labels.
problem Learning meaningful disentangled causal representations from observed data.
method ICM-VAE framework using flow-based diffeomorphic functions and causal disentanglement prior.
result Induces highly disentangled causal factors and improves robustness.
We analyze a family of methods for statistical causal inference from sample under the so-called Additive Noise Model. While most work on the subject has concentrated on establishing the soundness of the Additive Noise Model, the statistical consistency of the resulting inference methods has received little attention. W…
Hierarchical pretraining with slow-fast ODEs
problem Causal self-attention vs. slow-fast ODEs
method Instantiating fast-slow ODE formalism as a concrete neural network
result Equilibrium manifold x=φ(y) is exactly the master-equation (ME) stationary distribution 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.
Causal reasoning has been an indispensable capability for humans and other intelligent animals to interact with the physical world. In this work, we propose to endow an artificial agent with the capability of causal reasoning for completing goal-directed tasks. We develop learning-based approaches to inducing causal kn…
Transformer-based method for causal discovery with prior knowledge integration.
problem Complex nonlinear dependencies and spurious correlations in time series data.
method Multi-layer Transformer forecaster with gradient-based causal structure extraction and attention masking for prior knowledge integration.
result Significant improvement in causal discovery and causal lag estimation compared to state-of-the-art methods.
HKT improves sequence processing with multi-scale attention and kernel analysis.
problem Processing sequences at multiple scales with efficient attention mechanisms.
method Trainable causal downsampling and convex weights for level-specific score matrices.
result HKT achieves consistent gains over standard attention across various tasks.
New method identifies causal order without sparsity assumptions.
problem Causal order discovery in observational data.
method Sequential procedure to directly identify causal order.
result Direct identification of causal order without sparsity assumptions.
Transformers can predict new tokens based on any number of context tokens, approximating continuous mappings with fixed resources.
problem Handling an arbitrarily large number of context tokens in transformers.
method Mathematical analysis of transformer's expressivity using Wasserstein distance and continuous mappings.
result Deep transformers are universal and can approximate continuous in-context mappings to arbitrary precision, uniformly over compact token domains.
We review geometrical properties of a static spacetime (M,g), including geodesic completeness, causality, standard splittings, compact M, closed geodesics and geodesic connectedness. We pay special attention to the critical quadratic behavior at infinity of the coefficients β, β−1 (β=−g(K,K), being K a …
The machine learning community has recently devoted much attention to the problem of inferring causal relationships from statistical data. Most of this work has focused on uncovering connections among scalar random variables. We generalize existing methods to apply to collections of multi-dimensional random vectors, fo…
Extends causal discovery to group variables, improving performance in real-world applications.
problem Inferring cause-effect relationships from grouped data.
method Two-step approach: infer causal order and select models.
result Strong performance in simulations and real-world assembly line data.
We introduce a model for causal structure learning from multivariate functional data, even when graphs have cycles.
problem Discovering causal relationships from multivariate functional data with cycles.
method Functional linear structural equation model with a low-dimensional causal embedded space.
result The proposed model is causally identifiable under standard assumptions.
This survey explores causal inference in banking, finance, and insurance.
problem Explaining decisions in banking, finance, and insurance using causal inference.
method Categorizes 37 papers on causal inference applications in banking, finance, and insurance.
result Causal inference is still in its infancy in banking and insurance sectors.
New method for causal effect estimation with hidden confounders.
problem Estimating causal effects in the presence of hidden confounders.
method Singular value decomposition of a conditional expectation operator followed by saddle-point optimization.
result Our method outperforms existing methods on common benchmarks.
We propose a formulation of a Lorentzian quantum geometry based on the framework of causal fermion systems. After giving the general definition of causal fermion systems, we deduce space-time as a topological space with an underlying causal structure. Restricting attention to systems of spin dimension two, we derive th…
Causal autoregressive flows enable accurate causal inference and prediction.
problem Causal discovery and interventional predictions in machine learning.
method Autoregressive normalizing flows with fixed variable orderings.
result Causal models derived from autoregressive flows are identifiable and allow for accurate interventional and counterfactual predictions.
STAM learns important time steps and variables for multivariate time series prediction.
problem Accurate interpretation of multivariate time series predictions.
method Spatiotemporal attention mechanism (STAM) for multivariate time series modeling.
result STAM maintains state-of-the-art prediction accuracy with improved interpretability.
The discovery of non-linear causal relationship under additive non-Gaussian noise models has attracted considerable attention recently because of their high flexibility. In this paper, we propose a novel causal inference algorithm called least-squares independence regression (LSIR). LSIR learns the additive noise model…
This work integrates domain knowledge into A*-based causal discovery methods.
problem Efficiently incorporating domain knowledge into A*-based causal discovery methods.
method Integrates various types of domain knowledge into A*-based causal discovery methods, reducing the graph search space and improving computational gains.
result Small amounts of domain knowledge can dramatically speed up A*-based causal discovery and improve its performance and practicality.
Identification of causal direction between a causal-effect pair from observed data has recently attracted much attention. Various methods based on functional causal models have been proposed to solve this problem, by assuming the causal process satisfies some (structural) constraints and showing that the reverse direct…
New approach improves linear-time attention for language models.
problem Challenges of quadratic attention in long-sequence modelling, especially for discrete data.
method Reinterpreting linear attention through latent probabilistic graphical models, introducing asymmetric structure and recurrent parameterisation.
result Our model achieves competitive performance and outperforms existing linear attention variants on language modelling benchmarks.
Language models fail to process hallucinated responses, and this study diagnoses the failure.
problem Language models fail to process hallucinated responses, leading to over-concentration or diffuse attention.
method The study uses forced scoring of benchmark-labeled responses to compute attention shapes and analyze the symmetric component of the degree-normalized attention operator.
result The study proves that every transpose-invariant spectral diagnostic of the attention operator is orientation-blind and bounds the sensitivity of any Lipschitz diagnostic by the asymmetry coefficient \(G\).
Study examines causal properties of Finsler spacetimes with cone Killing vectors.
problem Characterize causality in Finsler spacetimes with specific Killing vectors.
method Explores the relationship between wind Riemannian structures and spacetimes with cone Killing vectors, focusing on Finsler-Kropina metrics.
result Characterizes causality properties using metric-type properties of Finslerian structures.
IntDC framework uncovers causal relationships from non-interventional data.
problem Detecting causal relationships in non-interventional complex systems.
method Interventional Embedding Entropy (IEE) for causal strength measurement.
result IEE accurately finds causal edges and quantifies causal strength robustly.
Measurement error in the observed values of the variables can greatly change the output of various causal discovery methods. This problem has received much attention in multiple fields, but it is not clear to what extent the causal model for the measurement-error-free variables can be identified in the presence of meas…