To date, most investigations on surprisal and entropy effects in reading have been conducted on the group level, disregarding individual differences. In this work, we revisit the predictive power of surprisal and entropy measures estimated from a range of language models (LMs) on data of human reading times as a measure of processing effort by incorporating information of language users' cognitive capacities. To do so, we assess the predictive power of surprisal and entropy estimated from generative LMs on reading data obtained from individuals who also completed a wide range of psychometric tests. Specifically, we investigate if modulating surprisal and entropy relative to cognitive scores increases prediction accuracy of reading times, and we examine whether LMs exhibit systematic biases in the prediction of reading times for cognitively high- or low-performing groups, revealing what type of psycholinguistic subject a given LM emulates. Our study finds that in most cases, incorporating cognitive capacities increases predictive power of surprisal and entropy on reading times, and that generally, high performance in the psychometric tests is associated with lower sensitivity to predictability effects. Finally, our results suggest that the analyzed LMs emulate readers with lower verbal intelligence, suggesting that for a given target group (i.e., individuals with high verbal intelligence), these LMs provide less accurate predictability estimates.
相關內容
Missing data often result in undesirable bias and loss of efficiency. These become substantial problems when the response mechanism is nonignorable, such that the response model depends on unobserved variables. It is necessary to estimate the joint distribution of unobserved variables and response indicators to manage nonignorable nonresponse. However, model misspecification and identification issues prevent robust estimates despite careful estimation of the target joint distribution. In this study, we modelled the distribution of the observed parts and derived sufficient conditions for model identifiability, assuming a logistic regression model as the response mechanism and generalised linear models as the main outcome model of interest. More importantly, the derived sufficient conditions are testable with the observed data and do not require any instrumental variables, which are often assumed to guarantee model identifiability but cannot be practically determined beforehand. To analyse missing data, we propose a new imputation method which incorporates verifiable identifiability using only observed data. Furthermore, we present the performance of the proposed estimators in numerical studies and apply the proposed method to two sets of real data: exit polls for the 19th South Korean election data and public data collected from the Korean Survey of Household Finances and Living Conditions.
We propose a numerical algorithm for the computation of multi-marginal optimal transport (MMOT) problems involving general probability measures that are not necessarily discrete. By developing a relaxation scheme in which marginal constraints are replaced by finitely many linear constraints and by proving a specifically tailored duality result for this setting, we approximate the MMOT problem by a linear semi-infinite optimization problem. Moreover, we are able to recover a feasible and approximately optimal solution of the MMOT problem, and its sub-optimality can be controlled to be arbitrarily close to 0 under mild conditions. The developed relaxation scheme leads to a numerical algorithm which can compute a feasible approximate optimizer of the MMOT problem whose theoretical sub-optimality can be chosen to be arbitrarily small. Besides the approximate optimizer, the algorithm is also able to compute both an upper bound and a lower bound for the optimal value of the MMOT problem. The difference between the computed bounds provides an explicit sub-optimality bound for the computed approximate optimizer. We demonstrate the proposed algorithm in three numerical experiments involving an MMOT problem that stems from fluid dynamics, the Wasserstein barycenter problem, and a large-scale MMOT problem with 100 marginals. We observe that our algorithm is capable of computing high-quality solutions of these MMOT problems and the computed sub-optimality bounds are much less conservative than their theoretical upper bounds in all the experiments.
Human capabilities in understanding visual relations are far superior to those of AI systems, especially for previously unseen objects. For example, while AI systems struggle to determine whether two such objects are visually the same or different, humans can do so with ease. Active vision theories postulate that the learning of visual relations is grounded in actions that we take to fixate objects and their parts by moving our eyes. In particular, the low-dimensional spatial information about the corresponding eye movements is hypothesized to facilitate the representation of relations between different image parts. Inspired by these theories, we develop a system equipped with a novel Glimpse-based Active Perception (GAP) that sequentially glimpses at the most salient regions of the input image and processes them at high resolution. Importantly, our system leverages the locations stemming from the glimpsing actions, along with the visual content around them, to represent relations between different parts of the image. The results suggest that the GAP is essential for extracting visual relations that go beyond the immediate visual content. Our approach reaches state-of-the-art performance on several visual reasoning tasks being more sample-efficient, and generalizing better to out-of-distribution visual inputs than prior models.
Consider a scenario where a large number of explanatory features targeting a response variable are analyzed, such that these features are partitioned into different groups according to their domain-specific structures. Furthermore, there may be several such partitions. Such multiple partitions may exist in many real-life scenarios. One such example is spatial genome-wide association studies. Researchers may not only be interested in identifying the features relevant to the response but also aim to determine the relevant groups within each partition. A group is considered relevant if it contains at least one relevant feature. To ensure the replicability of the findings at various resolutions, it is essential to provide false discovery rate (FDR) control for findings at multiple layers simultaneously. This paper presents a flexible approach that leverages various existing controlled selection procedures to generate more stable results with multilayer FDR control. The key contributions of our proposal are the development of a generalized e-filter that provides multilayer FDR control and the construction of a specific type of generalized e-values to evaluate feature importance. A primary application of our method is an improved version of Data Splitting (DS), called the eDS-filter. Furthermore, we combine the eDS-filter with the version of the group knockoff filter (gKF), resulting in a more flexible approach called the eDS+gKF filter. Simulation studies demonstrate that the proposed methods effectively control the FDR at multiple levels while maintaining or even improving power compared to other approaches. Finally, we apply the proposed method to analyze HIV mutation data.
In this paper, we apply quasi-Monte Carlo (QMC) methods with an initial preintegration step to estimate cumulative distribution functions and probability density functions in uncertainty quantification (UQ). The distribution and density functions correspond to a quantity of interest involving the solution to an elliptic partial differential equation (PDE) with a lognormally distributed coefficient and a normally distributed source term. There is extensive previous work on using QMC to compute expected values in UQ, which have proven very successful in tackling a range of different PDE problems. However, the use of QMC for density estimation applied to UQ problems will be explored here for the first time. Density estimation presents a more difficult challenge compared to computing the expected value due to discontinuities present in the integral formulations of both the distribution and density. Our strategy is to use preintegration to eliminate the discontinuity by integrating out a carefully selected random parameter, so that QMC can be used to approximate the remaining integral. First, we establish regularity results for the PDE quantity of interest that are required for smoothing by preintegration to be effective. We then show that an $N$-point lattice rule can be constructed for the integrands corresponding to the distribution and density, such that after preintegration the QMC error is of order $\mathcal{O}(N^{-1+\epsilon})$ for arbitrarily small $\epsilon>0$. This is the same rate achieved for computing the expected value of the quantity of interest. Numerical results are presented to reaffirm our theory.
We study average-reward Markov decision processes (AMDPs) and develop novel first-order methods with strong theoretical guarantees for both policy optimization and policy evaluation. Compared with intensive research efforts in finite sample analysis of policy gradient methods for discounted MDPs, existing studies on policy gradient methods for AMDPs mostly focus on regret bounds under restrictive assumptions, and they often lack guarantees on the overall sample complexities. Towards this end, we develop an average-reward stochastic policy mirror descent (SPMD) method for solving AMDPs with and without regularizers and provide convergence guarantees in terms of the long-term average reward. For policy evaluation, existing on-policy methods suffer from sub-optimal convergence rates as well as failure in handling insufficiently random policies due to the lack of exploration in the action space. To remedy these issues, we develop a variance-reduced temporal difference (VRTD) method with linear function approximation for randomized policies along with optimal convergence guarantees, and design an exploratory VRTD method that resolves the exploration issue and provides comparable convergence guarantees. By combining the policy evaluation and policy optimization parts, we establish sample complexity results for solving AMDPs under both generative and Markovian noise models. It is worth noting that when linear function approximation is utilized, our algorithm only needs to update in the low-dimensional parameter space and thus can handle MDPs with large state and action spaces.
Marginal structural models are a popular method for estimating causal effects in the presence of time-varying exposures. In spite of their popularity, no scalable non-parametric estimator exist for marginal structural models with multi-valued and time-varying treatments. In this paper, we use machine learning together with recent developments in semiparametric efficiency theory for longitudinal studies to propose such an estimator. The proposed estimator is based on a study of the non-parametric identifying functional, including first order von-Mises expansions as well as the efficient influence function and the efficiency bound. We show conditions under which the proposed estimator is efficient, asymptotically normal, and sequentially doubly robust in the sense that it is consistent if, for each time point, either the outcome or the treatment mechanism is consistently estimated. We perform a simulation study to illustrate the properties of the estimators, and present the results of our motivating study on a COVID-19 dataset studying the impact of mobility on the cumulative number of observed cases.
Recent advancements in feature representation and dimension reduction have highlighted their crucial role in enhancing the efficacy of predictive modeling. This work introduces TemporalPaD, a novel end-to-end deep learning framework designed for temporal pattern datasets. TemporalPaD integrates reinforcement learning (RL) with neural networks to achieve concurrent feature representation and feature reduction. The framework consists of three cooperative modules: a Policy Module, a Representation Module, and a Classification Module, structured based on the Actor-Critic (AC) framework. The Policy Module, responsible for dimensionality reduction through RL, functions as the actor, while the Representation Module for feature extraction and the Classification Module collectively serve as the critic. We comprehensively evaluate TemporalPaD using 29 UCI datasets, a well-known benchmark for validating feature reduction algorithms, through 10 independent tests and 10-fold cross-validation. Additionally, given that TemporalPaD is specifically designed for time series data, we apply it to a real-world DNA classification problem involving enhancer category and enhancer strength. The results demonstrate that TemporalPaD is an efficient and effective framework for achieving feature reduction, applicable to both structured data and sequence datasets. The source code of the proposed TemporalPaD is freely available as supplementary material to this article and at //www.healthinformaticslab.org/supp/.
The application of Shapley values to high-dimensional, time-series-like data is computationally challenging - and sometimes impossible. For $N$ inputs the problem is $2^N$ hard. In image processing, clusters of pixels, referred to as superpixels, are used to streamline computations. This research presents an efficient solution for time-seres-like data that adapts the idea of superpixels for Shapley value computation. Motivated by a forensic DNA classification example, the method is applied to multivariate time-series-like data whose features have been classified by a convolutional neural network (CNN). In DNA processing, it is important to identify alleles from the background noise created by DNA extraction and processing. A single DNA profile has $31,200$ scan points to classify, and the classification decisions must be defensible in a court of law. This means that classification is routinely performed by human readers - a monumental and time consuming process. The application of a CNN with fast computation of meaningful Shapley values provides a potential alternative to the classification. This research demonstrates the realistic, accurate and fast computation of Shapley values for this massive task
We investigate the prevalence of sample repetition in a Sequential Monte Carlo (SMC) method recently introduced for political redistricting.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191