Large-scale simulation models of complex socio-technical systems provide decision-makerswith high-fidelity testbeds inwhich policyinterventions canbe evaluated andwhat-if scenarios explored.

Federated learning (FL) is vulnerable to backdoor attacks due to its distributed computing nature. Existing defense solution usually requires larger amount of computation in either the training or testing phase, which limits their practicality in the resource-constrain scenarios. A more practical defense, i.e., neural network (NN) pruning based defense has been proposed in centralized backdoor setting. However, our empirical study shows that traditional pruning-based solution suffers \textit{poison-coupling} effect in FL, which significantly degrades the defense performance.This paper presents Lockdown, an isolated subspace training method to mitigate the poison-coupling effect. Lockdown follows three key procedures.

The coronavirus disease 2019 (COVID-19) global pandemic has led many countries to impose unprecedented lockdown measures in order to slow down the outbreak. Questions on whether governments have acted promptly enough, and whether lockdown measures can be lifted soon have since been central in public discourse. Data-driven models that predict COVID-19 fatalities under different lockdown policy scenarios are essential for addressing these questions, and for informing governments on future policy directions. To this end, this paper develops a Bayesian model for predicting the effects of COVID-19 containment policies in a global context -- we treat each country as a distinct data point, and exploit variations of policies across countries to learn country-specific policy effects. Our model utilizes a two-layer Gaussian process (GP) prior -- the lower layer uses a compartmental SEIR (Susceptible, Exposed, Infected, Recovered) model as a prior mean function with "country-and-policy-specific" parameters that capture fatality curves under different "counterfactual" policies within each country, whereas the upper layer is shared across all countries, and learns lower-layer SEIR parameters as a function of country features and policy indicators. Our model combines the solid mechanistic foundations of SEIR models (Bayesian priors) with the flexible data-driven modeling and gradient-based optimization routines of machine learning (Bayesian posteriors) -- i.e., the entire model is trained end-to-end via stochastic variational inference. We compare the projections of our model with other models listed by the Center for Disease Control (CDC), and provide scenario analyses for various lockdown and reopening strategies highlighting their impact on COVID-19 fatalities.

We use Erd os-Rényi Kernel (ERK) (Evci et al., 2020), an empirical (l 1) In the mask searching process, we use parameter's magnitude to guide the Following (Evci et al., 2020), we use Labels of poison samples are manipulated to the target label (e.g., a horse). " corresponds to be applicable while " " corresponds to be not applicable. BadNet is the earliest, and also the simplest backdoor attack first proposed in (Gu et al., DBA. DBA (Xie et al., 2019) is a backdoor attack specifically targeted on FL. Sinusoidal attack (Barni et al., 2019) shares a similar perspective with BadNet, The basic idea of Scaling (Bagdasaryan et al., 2020) is to enlarge the gradient update FixMask is an adaptive attack method specifically targeting Lockdown.

Federated learning (FL) is vulnerable to backdoor attacks due to its distributed computing nature.