Federated inference, in the form of one-shot federated learning, edge ensembles, or federated ensembles, has emerged as an attractive solution to combine predictions from multiple models. This paradigm enables each model to remain local and proprietary while a central server queries them and aggregates predictions. Yet, the robustness of federated inference has been largely neglected, leaving them vulnerable to even simple attacks. To address this critical gap, we formalize the problem of robust federated inference and provide the first robustness analysis of this class of methods. Our analysis of averaging-based aggregators shows that the error of the aggregator is small either when the dissimilarity between honest responses is small or the margin between the two most probable classes is large. Moving beyond linear averaging, we show that problem of robust federated inference with non-linear aggregators can be cast as an adversarial machine learning problem. We then introduce an advanced technique using the DeepSet aggregation model, proposing a novel composition of adversarial training and test-time robust aggregation to robustify non-linear aggregators. Our composition yields significant improvements, surpassing existing robust aggregation methods by 4.7 - 22.2% in accuracy points across diverse benchmarks.

Despite the widespread use of the data augmentation (DA) algorithm, the theoretical understanding of its convergence behavior remains incomplete. We prove the first non-asymptotic polynomial upper bounds on mixing times of three important DA algorithms: DA algorithm for Bayesian Probit regression (Albert and Chib, 1993, ProbitDA), Bayesian Logit regression (Polson, Scott, and Windle, 2013, LogitDA), and Bayesian Lasso regression (Park and Casella, 2008, Rajaratnam et al., 2015, LassoDA). Concretely, we demonstrate that with $\eta$-warm start, parameter dimension $d$, and sample size $n$, the ProbitDA and LogitDA require $\mathcal{O}\left(nd\log \left(\frac{\log \eta}{\epsilon}\right)\right)$ steps to obtain samples with at most $\epsilon$ TV error, whereas the LassoDA requires $\mathcal{O}\left(d^2(d\log d +n \log n)^2 \log \left(\frac{\eta}{\epsilon}\right)\right)$ steps. The results are generally applicable to settings with large $n$ and large $d$, including settings with highly imbalanced response data in the Probit and Logit regression. The proofs are based on the Markov chain conductance and isoperimetric inequalities. Assuming that data are independently generated from either a bounded, sub-Gaussian, or log-concave distribution, we improve the guarantees for ProbitDA and LogitDA to $\tilde{\mathcal{O}}(n+d)$ with high probability, and compare it with the best known guarantees of Langevin Monte Carlo and Metropolis Adjusted Langevin Algorithm. We also discuss the mixing times of the three algorithms under feasible initialization.

We consider latent structural versions of probit loss and ramp loss. We show that these surrogate loss functions are consistent in the strong sense that for any feature map (finite or infinite dimensional) they yield predictors approaching the infimum task loss achievable by any linear predictor over the given features. We also give finite sample generalization bounds (convergence rates) for these loss functions. These bounds suggest that probit loss converges more rapidly. However, ramp loss is more easily optimized on a given sample.

The shift between the training and testing distributions is commonly due to sample selection bias, a type of bias caused by non-random sampling of examples to be included in the training set. Although there are many approaches proposed to learn a classifier under sample selection bias, few address the case where a subset of labels in the training set are missing-not-at-random (MNAR) as a result of the selection process. In statistics, Greene's method formulates this type of sample selection with logistic regression as the prediction model. However, we find that simply integrating this method into a robust classification framework is not effective for this bias setting. In this paper, we propose BiasCorr, an algorithm that improves on Greene's method by modifying the original training set in order for a classifier to learn under MNAR sample selection bias. We provide theoretical guarantee for the improvement of BiasCorr over Greene's method by analyzing its bias. Experimental results on real-world datasets demonstrate that BiasCorr produces robust classifiers and can be extended to outperform state-of-the-art classifiers that have been proposed to train under sample selection bias.

With the scale of data growing every day, reducing the dimensionality (a.k.a. sketching) of high-dimensional data has emerged as a task of paramount importance. Relevant issues to address in this context include the sheer volume of data that may consist of categorical samples, the typically streaming format of acquisition, and the possibly missing entries. To cope with these challenges, the present paper develops a novel categorical subspace learning approach to unravel the latent structure for three prominent categorical (bilinear) models, namely, Probit, Tobit, and Logit. The deterministic Probit and Tobit models treat data as quantized values of an analog-valued process lying in a low-dimensional subspace, while the probabilistic Logit model relies on low dimensionality of the data log-likelihood ratios. Leveraging the low intrinsic dimensionality of the sought models, a rank regularized maximum-likelihood estimator is devised, which is then solved recursively via alternating majorization-minimization to sketch high-dimensional categorical data `on the fly.' The resultant procedure alternates between sketching the new incomplete datum and refining the latent subspace, leading to lightweight first-order algorithms with highly parallelizable tasks per iteration. As an extra degree of freedom, the quantization thresholds are also learned jointly along with the subspace to enhance the predictive power of the sought models. Performance of the subspace iterates is analyzed for both infinite and finite data streams, where for the former asymptotic convergence to the stationary point set of the batch estimator is established, while for the latter sublinear regret bounds are derived for the empirical cost. Simulated tests with both synthetic and real-world datasets corroborate the merits of the novel schemes for real-time movie recommendation and chess-game classification.