Search

We study the wild bootstrap inference for instrumental variable regressions in the framework of a small number of large clusters in which the number of clusters is viewed as fixed and the number of observations for each cluster diverges to infinity. We first show that the wild bootstrap Wald test, with or without using the cluster-robust covariance estimator, controls size asymptotically up to a small error as long as the parameters of endogenous variables are strongly identified in at least one of the clusters. Then, we establish the required number of strong clusters for the test to have power against local alternatives. We further develop a wild bootstrap Anderson-Rubin test for the full-vector inference and show that it controls size asymptotically up to a small error even under weak or partial identification in all clusters. We illustrate the good finite sample performance of the new inference methods using simulations and provide an empirical application to a well-known dataset about US local labor markets., Comment: This version only contains results for linear IV regressions. The results for IVQR in the previous versions will be put in a separate paper

Generalized zero-shot learning (GZSL) has achieved significant progress, with many efforts dedicated to overcoming the problems of visual-semantic domain gap and seen-unseen bias. However, most existing methods directly use feature extraction models trained on ImageNet alone, ignoring the cross-dataset bias between ImageNet and GZSL benchmarks. Such a bias inevitably results in poor-quality visual features for GZSL tasks, which potentially limits the recognition performance on both seen and unseen classes. In this paper, we propose a simple yet effective GZSL method, termed feature refinement for generalized zero-shot learning (FREE), to tackle the above problem. FREE employs a feature refinement (FR) module that incorporates \textit{semantic$\rightarrow$visual} mapping into a unified generative model to refine the visual features of seen and unseen class samples. Furthermore, we propose a self-adaptive margin center loss (SAMC-loss) that cooperates with a semantic cycle-consistency loss to guide FR to learn class- and semantically-relevant representations, and concatenate the features in FR to extract the fully refined features. Extensive experiments on five benchmark datasets demonstrate the significant performance gain of FREE over its baseline and current state-of-the-art methods. Our codes are available at https://github.com/shiming-chen/FREE ., Comment: ICCV 2021

Mode competition is conventionally regarded as an unwanted phenomenon that induces instability of laser emission. Here, we experimentally demonstrate that the random dynamics behavior of mode competition provides hidden information of the intracavity topology. We introduced the concept of self-assembled biophotonics networks, where human insulin protein with network topology was sandwiched in a Fabry-Perot (FP) cavity. The optical coupling between insulin fibrils and enables strong mode competition. Thanks to the optical feedback provided by the FP cavity, the interaction between the light and insulin network was enhanced. Subtle changes in the network will cause significant changes in laser mode dynamics. By recording the transverse mode dynamics, we reconstruct a graph that reveals the optical correlation in the insulin network. Our results indicate that the graph represents an intrinsic feature and can sensitivity respond to the stimulus. The unclonable and sensitively responsive property of the biophotonic lasing network can potentially be employed in applications in sensing and information encryption.

Recommender systems usually amplify the biases in the data. The model learned from historical interactions with imbalanced item distribution will amplify the imbalance by over-recommending items from the major groups. Addressing this issue is essential for a healthy ecosystem of recommendation in the long run. Existing works apply bias control to the ranking targets (e.g., calibration, fairness, and diversity), but ignore the true reason for bias amplification and trade-off the recommendation accuracy. In this work, we scrutinize the cause-effect factors for bias amplification, identifying the main reason lies in the confounder effect of imbalanced item distribution on user representation and prediction score. The existence of such confounder pushes us to go beyond merely modeling the conditional probability and embrace the causal modeling for recommendation. Towards this end, we propose a Deconfounded Recommender System (DecRS), which models the causal effect of user representation on the prediction score. The key to eliminating the impact of the confounder lies in backdoor adjustment, which is however difficult to do due to the infinite sample space of the confounder. For this challenge, we contribute an approximation operator for backdoor adjustment which can be easily plugged into most recommender models. Lastly, we devise an inference strategy to dynamically regulate backdoor adjustment according to user status. We instantiate DecRS on two representative models FM and NFM, and conduct extensive experiments over two benchmarks to validate the superiority of our proposed DecRS.