By training BPNN-D, a learned iterative operator that provably maintains many of the desirable properties of BP for any choice of the parameters, BPNNs learns to perform the task better than the original BP: it converges 1.7x faster on Ising models while providing tighter bounds.
The concept of query-based outlier in heterogeneous information networks is introduced, a query language is designed to facilitate users to specify outlier queries flexibly, a good outlier measure in heterogeneity networks is defined, and experiments show that following such a methodology, interesting outliers can be defined and uncovered flexibly and effectively in large heterogeneous networks.
A scheme where the family of hash functions is chosen adaptively, based on properties of the specific input formula is proposed, which leads to better lower bounds on existing benchmarks, with a median improvement factor of 213 over 1,198 propositional formulas.
It is shown that the weighted Rademacher complexity can be estimated by solving a randomly perturbed optimization problem, allowing us to derive high probability bounds on the size of any weighted set.
ADAPART uses an adaptive, iterative partitioning strategy over permutations to convert any upper bounding method for the permanent into one that satisfies a desirable `nesting' property over the partition used, and provides significant speedups over prior work.
This work presents a framework that combines a statistical inference technique known as conformal prediction with a simulator of robot/environment dynamics, in order to tune warning systems to provably achieve an ε false negative rate using as few as 1/ε data points.
It is shown that LSTMs outperform Kalman filtering for single target prediction by 2x, and a unique model for training two dependent LSTM to output a Gaussian distribution for a single target predictions to be used as input to multi-target tracking is presented.
This work introduces a framework that abstracts out the properties of recalibration problems under differential privacy constraints, and designs a novel recalIBration algorithm, accuracy temperature scaling, that outperforms prior work on private datasets and improves calibration on domain-shift benchmarks.
It is shown that the weighted Rademacher complexity can be estimated by solving a randomly perturbed optimization problem, allowing us to derive high-probability bounds on the size of any weighted set.
This paper precisely characterize these isomorphic properties of factor graphs and proposes two inference models: FactorEquivariant Neural Belief Propagation (FE-NBP and FE-GNN), a neural network that generalizes BP and respects each of the above properties.