Matching two sets of items, called set-to-set matching problem, is being recently raised. The difficulties of set-to-set matching over ordinary data matching lie in the exchangeability in 1) set-feature extraction and 2) set-matching score; the pair of sets and the items in each set should be exchangeable. In this paper, we propose a deep learning architecture for the set-to-set matching that overcomes the above difficulties, including two novel modules: 1) a cross-set transformation and 2) cross-similarity function. The former provides the exchangeable set-feature through interactions between two sets in intermediate layers, and the latter provides the exchangeable set matching through calculating the cross-feature similarity of items between two sets. We evaluate the methods through experiments with two industrial applications, fashion set recommendation, and group re-identification. Through these experiments, we show that the proposed methods perform better than a baseline given by an extension of the Set Transformer, the state-of-the-art set-input function.

Graphs and networks are a standard model for describing data or systems based on pairwise interactions. Oftentimes, the underlying relationships involve more than two entities at a time, and hypergraphs are a more faithful model. However, we have fewer rigorous methods that can provide insight from such representations. Here, we develop a computational framework for the problem of clustering hypergraphs with categorical edge labels --- or different interaction types --- where clusters corresponds to groups of nodes that frequently participate in the same type of interaction. Our methodology is based on a combinatorial objective function that is related to correlation clustering but enables the design of much more efficient algorithms. When there are only two label types, our objective can be optimized in polynomial time, using an algorithm based on minimum cuts. Minimizing our objective becomes NP-hard with more than two label types, but we develop fast approximation algorithms based on linear programming relaxations that have theoretical cluster quality guarantees. We demonstrate the efficacy of our algorithms and the scope of the model through problems in edge-label community detection, clustering with temporal data, and exploratory data analysis.

In federated learning systems, clients are autonomous in that their behaviors are not fully governed by the server. Consequently, a client may intentionally or unintentionally deviate from the prescribed course of federated model training, resulting in abnormal behaviors, such as turning into a malicious attacker or a malfunctioning client. Timely detecting those anomalous clients is therefore critical to minimize their adverse impacts. In this work, we propose to detect anomalous clients at the server side. In particular, we generate low-dimensional surrogates of model weight vectors and use them to perform anomaly detection. We evaluate our solution through experiments on image classification model training over the FEMNIST dataset. Experimental results show that the proposed detection-based approach significantly outperforms the conventional defense-based methods.

The high computational complexity associated with training deep neural networks limits online and real-time training on edge devices. This paper proposed an end-to-end training and inference scheme that eliminates multiplications by approximate operations in the log-domain which has the potential to significantly reduce implementation complexity. We implement the entire training procedure in the log-domain, with fixed-point data representations. This training procedure is inspired by hardware-friendly approximations of log-domain addition which are based on look-up tables and bit-shifts. We show that our 16-bit log-based training can achieve classification accuracy within approximately 1% of the equivalent floating-point baselines for a number of commonly used datasets.

We investigate online nonlinear regression with long short term memory (LSTM) based networks, which we refer to as LSTM-based online learning. For LSTM-based online learning, we introduce a highly efficient extended Kalman filter (EKF) based training algorithm with a theoretical convergence guarantee. Through simulations, we illustrate significant performance improvements achieved by our algorithm with respect to the conventional LSTM training methods. We particularly show that our algorithm provides very similar error performance with the EKF learning algorithm in 25-40 times shorter training time depending on the parameter size of the network.

Existing feature selection methods fail to properly account for interactions between features when evaluating feature subsets. In this paper, we attempt to remedy this issue by using orthogonal variance decomposition to evaluate features. The orthogonality of the decomposition allows us to directly calculate the total contribution of a feature to the output variance. Thus we obtain an efficient algorithm for feature evaluation which takes into account interactions among features. Numerical experiments demonstrate that our method accurately identifies relevant features and improves the accuracy of numerical models.

Within the last decade, neural network based predictors have demonstrated impressive - and at times super-human - capabilities. This performance is often paid for with an intransparent prediction process and thus has sparked numerous contributions in the novel field of explainable artificial intelligence (XAI). In this paper, we focus on a popular and widely used method of XAI, the Layer-wise Relevance Propagation (LRP). Since its initial proposition LRP has evolved as a method, and a best practice for applying the method has tacitly emerged, based on humanly observed evidence. We investigate - and for the first time quantify - the effect of this current best practice on feedforward neural networks in a visual object detection setting. The results verify that the current, layer-dependent approach to LRP applied in recent literature better represents the model's reasoning, and at the same time increases the object localization and class discriminativity of LRP.

A BSTRACT Since deep neural networks are over-parametrized, they may memorize noisy examples. We address such memorizing issue under the existence of annotation noise. From the fact that deep neural networks cannot generalize neighborhoods of the features acquired via memorization, we find that noisy examples do not consistently incur small losses on the network in the presence of perturbation. Based on this, we propose a novel training method called Learning with Ensemble Consensus (LEC) whose goal is to prevent overfitting noisy examples by eliminating them identified via consensus of an ensemble of perturbed networks. One of the proposed LECs, L TEC outperforms the current state-of-the-art methods on MNIST, CIFAR-10, and CIFAR-100 despite its efficient memory usage. 1 I NTRODUCTION Deep neural networks (DNNs) have shown excellent performance (Krizhevsky et al., 2012; He et al., 2016) on visual recognition datasets (Deng et al., 2009). However, it is difficult to obtain annotated datasets of such high quality in practice (Wang et al., 2018a). Even worse, DNNs may not generalize training data in the presence of noisy examples (Zhang et al., 2016). Therefore, there is an increasing demand for robust training methods. In general, DNNs trained on noisy datasets first generalize clean examples (Arpit et al., 2017).

Generative adversarial networks (GANs) have enjoyed much success in learning high-dimensional distributions. Learning objectives approximately minimize an $f$-divergence ($f$-GANs) or an integral probability metric (Wasserstein GANs) between the model and the data distribution using a discriminator. Wasserstein GANs enjoy superior empirical performance, but in $f$-GANs the discriminator can be interpreted as a density ratio estimator which is necessary in some GAN applications. In this paper, we bridge the gap between $f$-GANs and Wasserstein GANs (WGANs). First, we list two constraints over variational $f$-divergence estimation objectives that preserves the optimal solution. Next, we minimize over a Lagrangian relaxation of the constrained objective, and show that it generalizes critic objectives of both $f$-GAN and WGAN. Based on this generalization, we propose a novel practical objective, named KL-Wasserstein GAN (KL-WGAN). We demonstrate empirical success of KL-WGAN on synthetic datasets and real-world image generation benchmarks, and achieve state-of-the-art FID scores on CIFAR10 image generation.

A BSTRACT Learning disentangled representations that correspond to factors of variation in real-world data is critical to interpretable and human-controllable machine learning. Recently, concerns about the viability of learning disentangled representations in a purely unsupervised manner has spurred a shift toward the incorporation of weak supervision. However, there is currently no formalism that identifies when and how weak supervision will guarantee disentanglement. To address this issue, we provide a theoretical framework to assist in analyzing the disentanglement guarantees (or lack thereof) conferred by weak supervision when coupled with learning algorithms based on distribution matching. We empirically verify the guarantees and limitations of several weak supervision methods (restricted labeling, match-pairing, and rank-pairing), demonstrating the predictive power and usefulness of our theoretical framework. For example, in image generation, one might first generate the object identity and pose, and then build an image of this object accordingly. The goal of disentangled representation learning is to learn a representation where each dimensions of the representation measures a distinct factor of variation in the dataset (Bengio et al., 2013). Learning such representations that align with the underlying factors of variation may be critical to the development of machine learning models that are explainable or human-controllable (Gilpin et al., 2018; Lee et al., 2019; Klys et al., 2018). In recent years, disentanglement research has focused on the learning of such representations in an unsupervised fashion, using only independent samples from the data distribution without access to the true factors of variation (Higgins et al., 2017; Chen et al., 2018a; Kim & Mnih, 2018; Esmaeili et al., 2018). However, Locatello et al. (2019) demonstrated that many existing methods for the unsupervised learning of disentangled representations are brittle, requiring careful supervision-based hyperparameter tuning. To build robust disentangled representation learning methods that do not require large amounts of supervised data, recent work has turned to forms of weak supervision (Chen & Batmanghelich, 2019; Gabbay & Hoshen, 2019). Weak supervision can allow one to build models that have interpretable representations even when human labeling is challenging (e.g., hair style in face generation, or style in music generation). While existing methods based on weakly-supervised learning demonstrate empirical gains, there is no existing formalism for describing the theoretical guarantees conferred by different forms of weak supervision (Kulkarni et al., 2015; Reed et al., 2015; Bouchacourt et al., 2018).