Abstract-In many audio processing tasks, such as source separation, denoising or compression, it is crucial to construct realistic and flexible models to capture the physical properties of audio signals. This can be accomplished in the Bayesian framework through the use of appropriate prior distributions. In this paper, we describe a class of prior models called Gamma Markov random fields (GMRFs) to model the sparsity and the local dependency of the energies (i.e., variances) of time-frequency expansion coefficients. A GMRF model describes a non-normalised joint distribution over unobserved variance variables, where given the field the actual source coefficients are independent. Our construction ensures a positive coupling between the variance variables, so that signal energy changes smoothly over both axes to capture the temporal and spectral continuity. The coupling strength is controlled by a set of hyperparameters. Inference on the overall model is convenient because of the conditional conjugacy of all of the variables in the model, but automatic optimization of hyperparameters is crucial to obtain better fits. The marginal likelihood of the model is not available because of the intractable normalizing constant of GMRFs. In this paper, we optimize the hyperparameters of our GMRF-based audio model using contrastive divergence and compare this method to alternatives such as score matching and pseudolikelihood maximization where applicable. We present the performance of the GMRF models in denoising and single-channel source separation problems in completely blind scenarios, where all the hyperparameters are jointly estimated given only audio data.

No abstract

No abstract

BackgroundUsers of a personalised recommendation system face a dilemma: recommendations can be improved by learning from data, but only if other users are willing to share their private information. Good personalised predictions are vitally important in precision medicine, but genomic information on which the predictions are based is also particularly sensitive, as it directly identifies the patients and hence cannot easily be anonymised. Differential privacy has emerged as a potentially promising solution: privacy is considered sufficient if presence of individual patients cannot be distinguished. However, differentially private learning with current methods does not improve predictions with feasible data sizes and dimensionalities.ResultsWe show that useful predictors can be learned under powerful differential privacy guarantees, and even from moderately-sized data sets, by demonstrating significant improvements in the accuracy of private drug sensitivity prediction with a new robust private regression method. Our method matches the predictive accuracy of the state-of-the-art non-private lasso regression using only 4x more samples under relatively strong differential privacy guarantees. Good performance with limited data is achieved by limiting the sharing of private information by decreasing the dimensionality and by projecting outliers to fit tighter bounds, therefore needing to add less noise for equal privacy.ConclusionsThe proposed differentially private regression method combines theoretical appeal and asymptotic efficiency with good prediction accuracy even with moderate-sized data. As already the simple-to-implement method shows promise on the challenging genomic data, we anticipate rapid progress towards practical applications in many fields.ReviewersThis article was reviewed by Zoltan Gaspari and David Kreil.Electronic supplementary materialThe online version of this article (10.1186/s13062-017-0203-4) contains supplementary material, which is available to authorized users.