… approach to finding the maximum a posteriori probability of a set of random variables, given
some prior probability distribution … assume that these probability distributions are Gaussian. …

… are automatically chosen to approximate a given … scalable for massive datasets. Crucially,
the inference algorithms can also be parallelized to take full advantage of large distributed-…

… algorithm and our methods for estimating distribution parameters. … the statistical information
of the sensor data, it suffices if the aggregation algorithms return the probability distribution of …

… Scalable Kernel Composition (SKC), a scalable extension of CKS, to push the boundaries of
automated interpretable statistical … We later use the Nyström approximation to give an upper …

… approximate distributions of interest by carefully combining sequential Monte Carlo methods
with methodology … in order for it to recover a probability distribution of interest. Section 3 is …

… rithms that are scalable to very large data sets and that also have good statistical or …
computation—either in the sense of approximation algorithms in theoretical computer science or in …

… Stein’s method compares probability distributions through … method and statistics. The topics
we discuss include tools to benchmark and compare sampling methods such as approximate …

… of accuracy and computational scalability for estimating output statistics, compared to existing
uncertainty sampling approaches, such as stochastic response surface method and SG [19…

… tools to build scalable solvers for linear systems [313, 318, 317, 319], Gaussian sampling
[86], and geometric median [98]. Scalable techniques for PageRank approximation [65] have …

… When there are no or few Byzantine machines, we see the usual scaling 1 … We consider
the problem of estimating the mean µ of some random variable z ∼ D, which is equivalent to …