Filters








67 Hits in 6.3 sec

Compressed Sensing with Upscaled Vector Approximate Message Passing [article]

Nikolajs Skuratovs, Michael Davies
2022 arXiv   pre-print
We propose a rigorous method for correcting and tuning CG withing CG-VAMP to achieve a stable and efficient reconstruction.  ...  The Recently proposed Vector Approximate Message Passing (VAMP) algorithm demonstrates a great reconstruction potential at solving compressed sensing related linear inverse problems.  ...  We consider the largescale compressed sensing scenario where M < N with the ratio M N = δ = O (1) . Additionally, we are interested in the cases where the operator A might be ill-conditioned.  ... 
arXiv:2011.01369v3 fatcat:g65t6tdqdreevjdlgxvqljku7u

Double-Sided Information Aided Temporal-Correlated Massive Access [article]

Weifeng Zhu, Meixia Tao, Yunfeng Guan
2022 arXiv   pre-print
The DSI is extracted from the estimation results in a sliding window that contains the target detection frame and its previous and next frames.  ...  Motivated by that the device activity at each frame is correlated to not only its previous frame but also its next frame, we propose a double-sided information (DSI) aided joint activity detection and  ...  The authors are with the Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai 200240, China (e-mail: {wf.zhu, mxtao, yfguan69}@sjtu.edu.cn).  ... 
arXiv:2205.07494v1 fatcat:xgnqo4julfbshcqmwvjsglqviu

Deeptime: a Python library for machine learning dynamical models from time series data [article]

Moritz Hoffmann, Martin Scherer, Tim Hempel, Andreas Mardt, Brian de Silva, Brooke E. Husic, Stefan Klus, Hao Wu, Nathan Kutz, Steven L. Brunton, Frank Noé
2021 arXiv   pre-print
The library is largely compatible with scikit-learn, having a range of Estimator classes for these different models, but in contrast to scikit-learn also provides deep Model classes, e.g. in the case of  ...  In the physical sciences, structures such as metastable and coherent sets, slow relaxation processes, collective variables dominant transition pathways or manifolds and channels of probability flow can  ...  Variational approach for Markov processes (VAMP) .  ... 
arXiv:2110.15013v1 fatcat:6kjojkzt4bfwnaoxfmzwepno7a

Stochastic-Expert Variational Autoencoder for Collaborative Filtering

Yoon-Sik Cho, Min-hwan Oh
2022 Proceedings of the ACM Web Conference 2022  
Motivated by the recent successes of deep generative models used for collaborative filtering, we propose a novel framework of VAE for collaborative filtering using multiple experts and stochastic expert  ...  In our method, individual experts are sampled stochastically at each user-item interaction which can effectively utilize the variability among multiple experts.  ...  σ ϕ (•) are the variational parameters for Kdimensional mean and variance respectively as in the previous VAE models [10, 14] .  ... 
doi:10.1145/3485447.3512120 fatcat:km4bnt35inemfjvrpt2vrougx4

SignalNet: A Low Resolution Sinusoid Decomposition and Estimation Network [article]

Ryan Dreifuerst, Robert W. Heath Jr
2022 arXiv   pre-print
The detection and estimation of sinusoids is a fundamental signal processing task for many applications related to sensing and communications.  ...  Similarly, low resolution sampling in imaging and spectrum sensing allows for efficient data collection.  ...  We also investigate EM-VAMP, a message passing algorithm, as a method for determining the model order in a compressed sensing form.  ... 
arXiv:2106.05490v2 fatcat:5jwhwxys65amxbrdfb2rqij6yq

Machine Learning in Molecular Dynamics Simulations of Biomolecular Systems [article]

Christopher Kolloff, Simon Olsson
2022 arXiv   pre-print
This information allows mechanistic descriptions of molecular mechanisms, enables a quantitative comparison with experiments, and facilitates their rational design.  ...  Its success has also led to several synergies with molecular dynamics (MD) simulations, which we use to identify and characterize the major metastable states of molecular systems.  ...  Acknowledgement This work was supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation.  ... 
arXiv:2205.03135v1 fatcat:35jazyvdqvgzxmjjcl5gawi2xi

Music Expectation by Cognitive Rule-Mapping

Eugene Narmour
2000 Music Perception  
Rhythmic augmentation and diminution, by contrast, rely on multiplication and division. The examples suggest numerous hypotheses for experimental research.  ...  Rhythmic augmentation and diminution, by contrast, rely on multiplication and division. The examples suggest numerous hypotheses for experimental research.  ...  Listeners can absorb and project this formal "compression," even though Beethoven's music requires other kinds of attentional processing.  ... 
doi:10.2307/40285821 fatcat:evpiaasnvrcgdizb5cw772t6jy

Patch-Based Image Restoration using Expectation Propagation [article]

Dan Yao and Stephen McLaughlin and Yoann Altmann
2021 arXiv   pre-print
Experiments conducted for denoising, inpainting and deconvolution problems with Gaussian and Poisson noise illustrate the potential benefits of such flexible approximate Bayesian method for uncertainty  ...  Moreover, imposing structural constraints on the covariance matrices of these densities allows for greater scalability and distributed computation.  ...  The proposed EP algorithm was performed without parallel implementation, i.e., the J image patches and r experts are processed using a sequential for loop, and execution times are reported for comparison  ... 
arXiv:2106.15327v2 fatcat:2ict7e4upbfdllo2zo4rhsunwi

Comparison of Anomaly Detectors: Context Matters [article]

Vít Škvára, Jan Franců, Matěj Zorek, Tomáš Pevný, Václav Šmídl
2021 arXiv   pre-print
All our code and results are available for download.  ...  The objective of this comparison is twofold: to compare anomaly detection methods of various paradigms with focus on deep generative models, and identification of sources of variability that can yield  ...  Second, we study hyperparameter selection for two individual methods, variational autoencoder family and OC-SVM. 1) Impact of the number of anomalies in the validation set: The process of hyperparameter  ... 
arXiv:2012.06260v3 fatcat:min2zullxrbupf36o6uoeauwhi

Solving Sparse Linear Inverse Problems in Communication Systems: A Deep Learning Approach With Adaptive Depth [article]

Wei Chen, Bowen Zhang, Shi Jin, Bo Ai, Zhangdui Zhong
2020 arXiv   pre-print
We conduct experiments using both synthetic data and applications including random access in massive MTC and massive MIMO channel estimation, and the results demonstrate the improved efficiency for the  ...  However, it ignores a key character in traditional iterative algorithms, where the number of iterations required for convergence changes with varying sparsity levels.  ...  Applying compressive sensing with the sparse linear inverse problem leads to improve energy efficiency [2] , [3] .  ... 
arXiv:2010.15376v1 fatcat:xp7a3447nzd5ri7uiah22tb25e

Tonality Estimation In Electronic Dance Music: A Computational And Musically Informed Examination

Ángel Faraldo, Sergi Jordà, Perfecto Herrera
2018 Zenodo  
Based on this corpus, I propose the creation of more open-ended key labels, accounting for other modal practises and ambivalent tonal configurations.  ...  Last, I describe my own key finding methods, adapting existing models to the musical idiosyncrasies and tonal distributions of electronic dance music, with new statistical key profiles derived from the  ...  Some of the things I have learnt will stay with me for a long time, and they will hopefully manifest transmuted into different realities, knowledge, music and research.  ... 
doi:10.5281/zenodo.1154586 fatcat:fqd24dcqlvhflcsluwpja2x4zu

Music Information Retrieval: Recent Developments and Applications

Markus Schedl, Emilia Gómez, Julián Urbano
2014 Foundations and Trends in Information Retrieval  
We first elaborate on well-established and proven methods for feature extraction and music indexing, from both the audio signal and contextual data sources about music items, such as web pages or collaborative  ...  Subsequently, we review current work on user analysis and modeling in the context of music recommendation and retrieval, addressing the recent trend towards user-centric and adaptive approaches and systems  ...  The authors would further like to thank Masataka Goto, Mohamed Sordo, and Edith Law for granting permission to include their visual material in this survey, and Juan J.  ... 
doi:10.1561/1500000042 fatcat:c5tjdcy3xrfqvp6isnktbr6lpy

Deeptime: a Python library for machine learning dynamical models from time series data

Moritz Hoffmann, Martin Scherer, Tim Hempel, Andreas Mardt, Brian De Silva, Brooke E. Husic, Stefan Klus, Hao Wu, Nathan Kutz, Frank Noé, Universitätsbibliothek Der FU Berlin
2022
The library is largely compatible with scikit-learn, having a range of Estimator classes for these different models, but in contrast to scikit-learn also provides deep Model classes, e.g. in the case of  ...  In the physical sciences, structures such as metastable and coherent sets, slow relaxation processes, collective variables, dominant transition pathways or manifolds and channels of probability flow can  ...  Part of this research was performed while M H, A M, B E H, S K, H W, N K, S L B, and F N were visiting the Institute for Pure and Applied Mathematics (IPAM), which is supported by the National Science  ... 
doi:10.17169/refubium-33191 fatcat:sjqdgbupjrdc5doeuidezfehda

A Deterministic and Generalized Framework for Unsupervised Learning with Restricted Boltzmann Machines [article]

Eric W. Tramel and Marylou Gabrié and Andre Manoel and Francesco Caltagirone and Florent Krzakala
2017 arXiv   pre-print
In this work, we derive a deterministic framework for the training, evaluation, and use of RBMs based upon the Thouless-Anderson-Palmer (TAP) mean-field approximation of widely-connected systems with weak  ...  with sampling.  ...  V, solutions of the TAP GRBM free energy can be found by a fixed-point iteration, as shown in Alg. 1, which bears much resemblance to the AMP iteration derived in the context of compressed sensing [59  ... 
arXiv:1702.03260v2 fatcat:ngxwzpu77bghndibp6ny4conzm

From Heuristics-Based To Data-Driven Audio Melody Extraction

Juan J. Bosch, Emilia Gómez
2017 Zenodo  
This thesis investigates the benefits of exploiting knowledge automatically derived from data for audio melody extraction, by combining digital signal processing and machine learning methods.  ...  We extend the scope of melody extraction research by working with a varied dataset and multiple definitions of melody.  ...  This process takes place by two inter-dependent processes: sequential grouping, which senses data over time, and simultaneous grouping, which groups components of sounds which arrive at the same time.  ... 
doi:10.5281/zenodo.1120333 fatcat:4725nf75x5ds5glhoy3n2nfkea
« Previous Showing results 1 — 15 out of 67 results