Abstract:Generalized additive models (GAM) have been successfully applied to high dimensional data analysis. However, most existing methods cannot simultaneously estimate the link function, the component functions and the variable interaction. To alleviate this problem, we propose a new sparse additive model, named generalized sparse additive model with unknown link function (GSAMUL), in which the component functions are estimated by B-spline basis and the unknown link function is estimated by a multi-layer perceptron (MLP) network. Furthermore, $\ell_{2,1}$-norm regularizer is used for variable selection. The proposed GSAMUL can realize both variable selection and hidden interaction. We integrate this estimation into a bilevel optimization problem, where the data is split into training set and validation set. In theory, we provide the guarantees about the convergence of the approximate procedure. In applications, experimental evaluations on both synthetic and real world data sets consistently validate the effectiveness of the proposed approach.
Abstract:Dense waveguide arrays with low crosstalk and ultra-broadband remain a vital issue for chip-scale integrated photonics. However, the sub-wavelength regime of such devices has not been adequately explored in practice. Herein, we propose the advanced waveguide superlattices leveraging the artificial gauge field mechanism. This approach achieves remarkable -24 dB crosstalk suppression with an ultra-broadband bandwidth, experimentally demonstrated over 500 nm, in silicon nitride waveguides. Moreover, the 112 Gbit/s signal encoded per channel of ultra-compact circuits with a bit error rate below the 7% forward error correction limit verified the capability for high-speed on-chip transmission. This design is compatible with metal back end-of-the-line (BEOL) processes and can be readily transferred to other platforms. Thus it holds great promise for significant reduction in on-chip footprint and cost in large-scale integrated photonics, and salient enhancement in the performance of a wide range of active and passive photonic devices and systems.
Abstract:Deep neural networks have demonstrated promising performance on image recognition tasks. However, they may heavily rely on confounding factors, using irrelevant artifacts or bias within the dataset as the cue to improve performance. When a model performs decision-making based on these spurious correlations, it can become untrustable and lead to catastrophic outcomes when deployed in the real-world scene. In this paper, we explore and try to solve this problem in the context of skin cancer diagnosis. We introduce a human-in-the-loop framework in the model training process such that users can observe and correct the model's decision logic when confounding behaviors happen. Specifically, our method can automatically discover confounding factors by analyzing the co-occurrence behavior of the samples. It is capable of learning confounding concepts using easily obtained concept exemplars. By mapping the black-box model's feature representation onto an explainable concept space, human users can interpret the concept and intervene via first order-logic instruction. We systematically evaluate our method on our newly crafted, well-controlled skin lesion dataset and several public skin lesion datasets. Experiments show that our method can effectively detect and remove confounding factors from datasets without any prior knowledge about the category distribution and does not require fully annotated concept labels. We also show that our method enables the model to focus on clinical-related concepts, improving the model's performance and trustworthiness during model inference.