arXiv research

Chronological Causal Bandits (CCB) tackles dynamic causal decision-making.

Recently ({\em Class. Quant. Grav.} {\bf 20} 625-664) the concept of {\em causal mapping} between spacetimes --essentially equivalent in this context to the {\em chronological map} one in abstract chronological spaces--, and the related notion of {\em causal structure}, have been introduced as new tools to study causal…

The paper examines how timing of observations affects causal discovery methods.

Researchers examine various causal structures for spacetimes with continuous metrics.

Recently, the intervention calculus when the DAG is absent (IDA) method was developed to estimate lower bounds of causal effects from observational high-dimensional data. Originally it was introduced to assess the effect of baseline biomarkers which do not vary over time. However, in many clinical settings, measurement…

Causal classification of three Misner-type spacetimes.

Study causal structure of warped spacetimes using novel pre-length spaces.

Reconstruct spacetime from order and number of points.

Irrespective of local conditions imposed on the metric, any extendible spacetime U has a maximal extension containing no closed causal curves outside the chronological past of U. We prove this fact and interpret it as impossibility (in classical general relativity) of the time machines, insofar as the latter are define…

New algorithms for causal bandits without knowing the graph structure.

New model tackles causal bandits with dependent variables.

The paper establishes criteria for spacetime inextendibility using asymptotic volume-distance-ratio analysis.

We demonstrate the breakdown of several fundamentals of Lorentzian causality theory in low regularity. Most notably, chronological futures (defined naturally using locally Lipschitz curves) may be non-open, and may differ from the corresponding sets defined via piecewise $C^1$-curves. By refining the notion of a causal…

This paper identifies the minimal set of nodes for optimal conditional interventions in causal bandits.

Bandit is a framework for designing sequential experiments. In each experiment, a learner selects an arm $A \in \mathcal{A}$ and obtains an observation corresponding to $A$. Theoretically, the tight regret lower-bound for the general bandit is polynomial with respect to the number of arms $|\mathcal{A}|$. This makes ba…

On the Geroch-Kronheimer-Penrose future completion $IP(X)$ of a spacetime $X$, there are two frequently used topologies. We systematically examine $τ_+$, the stronger (metrizable) of them, which is the coarsest causally continuous topology, obtaining a variety of novel results, among them a complete characterization of…

We study the problem of using causal models to improve the rate at which good interventions can be learned online in a stochastic environment. Our formalism combines multi-arm bandits and causal inference to model a novel type of bandit feedback that is not exploited by existing approaches. We propose a new algorithm t…

The paper explores how to apply causal knowledge across different datasets to improve learning.

The paper tackles bandit problems with biased offline data by using causal methods.

This work explores adaptive strategies for multi-armed bandits with causal structure, achieving optimal regret bounds.

This work explores the idea of a causal contextual multi-armed bandit approach to automated marketing, where we estimate and optimize the causal (incremental) effects. Focusing on causal effect leads to better return on investment (ROI) by targeting only the persuadable customers who wouldn't have taken the action orga…

New algorithm learns causal graph to minimize regret in bandits without full structure.

New algorithm identifies best intervention without graph knowledge.

Clarifies definitions of global hyperbolicity in various spaces.

New algorithm finds causal parent nodes without graph learning.

Algorithm BGLM-OFU minimizes regret in combinatorial causal bandits with binary models.

New algorithm reduces regret in combinatorial causal bandits without graph structure.

Adaptive algorithm reduces regret in causal bandits.

We study how to learn optimal interventions sequentially given causal information represented as a causal graph along with associated conditional distributions. Causal modeling is useful in real world problems like online advertisement where complex causal mechanisms underlie the relationship between interventions and …

The paper tackles policy learning in dynamic environments using causal methods.

Study adapts combinatorial semi-bandit for piecewise stationary, causally related rewards.

This paper reviews causal inference methods for time series data.

IMPaCT improves node classification in chronological split temporal graphs.

The paper tackles causal bandits for SEMs, proposing algorithms that avoid estimating $2^N$ reward distributions.

Chronologically consistent models maintain accuracy with time-restricted data.

DFPV improves PCL for confounded bandit policy evaluation.

Defines doubling conditions for Lorentzian spaces linked to curvature bounds.

Combines offline causal inference and online bandit learning for better decision-making.

A new algorithm tackles delayed combinatorial semi-bandit with causal relations.

Corrects pseudo log-likelihood method issues in various applications.

Recently, a new viewpoint on the classical c-boundary in Mathematical Relativity has been developed, the relations of this boundary with the conformal one and other classical boundaries have been analyzed, and its computation in some classes of spacetimes, as the standard stationary ones, has been carried out. In the p…

We show that when a spacetime $\mathcal{M}(=M \cup \partial M)$ is globally hyperbolic with (possibly empty) smooth timelike boundary $\partial M$, a metrizable topology, the closed limit topology (CLT) introduced by F. Hausdorff himself in the 1950's in set theory, can be advantageously adopted on the Geroch-Kronheime…

The paper tackles causal bandits with unknown SCMs and soft interventions, providing upper and lower bounds on regret.

In a recent paper, Eichmair, Galloway and Pollack have proved a Gannon-Lee-type singularity theorem based on the existence of marginally outer trapped surfaces (MOTS) on noncompact initial data sets for globally hyperbolic spacetimes. However, one might wonder whether the corresponding incomplete geodesics could still …

New algorithm reduces regret in CBs with time-varying models.

The paper characterizes global hyperbolicity in Lorentzian manifolds without relying on manifold topology.

Recent results on the maximization of the charged-particle action I in a globally hyperbolic spacetime are discussed and generalized. We focus on the maximization of I over a given causal homotopy class C of curves connecting two causally related events x_0 <= x_1. Action I is proved to admit a maximum on C, and also o…

New algorithms for efficient causal interventions with budget constraints and without constraints.