The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The minimization of empirical risks over finite sample sizes is an important problem in large-scale machine learning. A variety of algorithms has been proposed in the literature to alleviate the computational burden per iteration at the expense of convergence speed and accuracy. Many of these approaches can be interpreted as stochastic gradient descent algorithms, where data is sampled from particular...
We propose a new methodology for explaining the predictions of black box classifiers. We use the motivating paradigm that predictive performance is of primary importance but human analysts (e.g., in fraud detection) desire a classifier's predictions to be augmented with useful explanations. To be truly general and principled, we derive a scoring system for finding explanations based on formal requirements...
Both boosting and deep stacking sequentially train their units taking into account the outputs of the previously trained learners. This parallelism suggests that it exists the possibility of getting some advantages by combining these techniques, i.e., emphasis and injection, in appropiate manners. In this paper, we propose a first mode for such a combination by simultaneously applying a general and...
Many biological monitoring projects rely on acoustic detection of birds. Despite increasingly large datasets, this detection is often manual or semi-automatic, requiring manual tuning/postprocessing. We review the state of the art in automatic bird sound detection, and identify a widespread need for tuning-free and species-agnostic approaches. We introduce new datasets and an IEEE research challenge...
Training deep neural networks requires a large amount of memory, making very deep neural networks difficult to fit on accelerator memories. In order to overcome this limitation, we present a method to reduce the amount of memory for training a deep neural network. The method enables to suppress memory increase during the backward pass, by reusing the memory regions allocated for the forward pass....
A patient's estimated glomerular filtration rate (eGFR) can provide important information about disease progression and kidney function. Traditionally, an eGFR time series is interpreted by a human expert labelling it as stable or unstable. While this approach works for individual patients, the time consuming nature of it precludes the quick evaluation of risk in large numbers of patients. However,...
We consider the problem of learning graphs in a sparse multiclass support vector machines framework. For such a problem, sparse graph penalty is useful to select the significant features and interpret the results. Classical ℓ1-norm learns a sparse solution without considering the structure between the features. In this paper, a structural knowledge is encoded as directed acyclic graph and a graph...
We propose a tuning-free Bayesian approach to learn a set of sparse graphical models, in which adjacent graphs share similar structures. This model can be applied to estimating dynamic networks that evolve smoothly with regard to a covariate (e.g., time). Specifically, a novel structured spike and slab prior is constructed. This prior allows time-varying sparsity pattern by smoothing the spike probabilities...
Nonlinear acoustic echo cancellation (NAEC) aims at estimating both the acoustic impulse response and the nonlinearities affecting the desired signal. Both the modeling processes show behaviors of sparse nature from an energy point of view. In this paper, we propose an adaptive NAEC algorithm that takes advantage of such sparsity behaviors to improve echo cancellation performance. The proposed scheme...
We propose a new approach to time-frequency mask generation for real-time multichannel speech separation. Whereas conventional approaches select the strongest source in each time-frequency bin, we perform a binary hypothesis test to determine whether a target source is present or not. We derive a generalized likelihood ratio test and extend it to underdetermined mixtures by aggregating the outputs...
We introduce a new method for estimating the regime-switching stochastic volatility models from the historical prices. Our methodology is based on a novel version of the assumed density filter (ADF). We estimate the switching model by maximizing the quasi-likelihood function of our ADF. The simulation experiments show the efficiency of our method. Then we analyze different market price histories for...
In this paper we present the application of ensemble learning to epileptic seizure detection problem. We propose a robust learning framework to mitigate class imbalance in large CHB-MIT (982 hrs) scalp EEG dataset. The algorithm being used is RUSBoost which is a hybrid data sampling and boosting technique designed especially for skewed classes. The data that is being used in this study has severe...
Matrix factorization is a key component of collaborative filtering-based recommendation systems because it allows us to complete sparse user-by-item ratings matrices under a low-rank assumption that encodes the belief that similar users give similar ratings and that similar items garner similar ratings. This paradigm has had immeasurable practical success, but it is not the complete story for understanding...
The computational complexity of kernel methods grows at least quadratically with respect to the training size and hence low rank kernel approximation techniques are commonly used. One of the most popular approximations is constructed by sub-sampling the training data. In this paper, we present a sampling algorithm called Enhanced Distance Subset Approximation (EDSA) based on a novel kernel function...
Auditory-evoked noninvasive electroencephalography (EEG) based brain-computer interfaces (BCIs) could be useful for improved hearing aids in the future. This manuscript investigates the role of frequency and spatial features of audio signal in EEG activities in an auditory BCI system with the purpose of detecting the attended auditory source in a cocktail party setting. A cross correlation based feature...
The infinite relational model (IRM) is a Bayesian nonparametric stochastic block model; a generative model for random networks parameterized for uni-partite undirected networks by a partition of the node set and symmetric matrix of inter-partion link probabilities. The prior for the node clusters is the Chinese restaurant process, and the link probabilities are, in the most simple setting, modeled...
Automatic identification of jump Markov systems (JMS) is known to be an important but difficult problem. In this work, we propose a new algorithm for the unsupervised estimation of parameters in a class of linear JMS called “conditionally Gaussian pairwise Markov switching models” (CGPMSMs), which extends the family of classic “conditionally Gaussian linear state-space models” (CGLSSMs). The method...
We present a dictionary learning algorithm that aims to reduce the size of the dictionary to a parsimonious value during the learning process. The sparse coding step uses a weighted Orthogonal Matching Pursuit favoring atoms that enter more representations. The dictionary update step optimizes a regularized error, encouraging the apparition of zero rows in the representation matrix; the corresponding...
There arises the need in many wireless network applications to infer and track different models of interest. Some nodes in the network are informed, where they observe the different models and send information to the uninformed ones. Each uninformed node responds to one informed node and joins its group. In this work, we suggest an adaptive and distributed clustering and partitioning approach that...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.