Filters








13 Hits in 0.96 sec

Masking Strategies for Image Manifolds [article]

Hamid Dadkhahi, Marco F. Duarte
2016 arXiv   pre-print
Masking Strategies for Image Manifolds Hamid Dadkhahi and Marco F.  ...  Dadkhahi physical system or event governed by a few continuousvalued parameters.  ... 
arXiv:1606.04618v1 fatcat:bjahqbxd3bcnfcxsy2opvlos2i

Alternating Linear Bandits for Online Matrix-Factorization Recommendation [article]

Hamid Dadkhahi, Sahand Negahban
2018 arXiv   pre-print
Dadkhahi College of Information and Computer Sciences University of Massachusetts Amherst  ...  Alternating Linear Bandits for Online Matrix-Factorization Recommendation Hamid  ... 
arXiv:1810.09401v1 fatcat:h4mm75e4sngrjikkgxg33y4f5a

Learning Shallow Detection Cascades for Wearable Sensor-Based Mobile Health Applications [article]

Hamid Dadkhahi, Nazir Saleheen, Santosh Kumar, Benjamin Marlin
2016 arXiv   pre-print
The field of mobile health aims to leverage recent advances in wearable on-body sensing technology and smart phone computing capabilities to develop systems that can monitor health states and deliver just-in-time adaptive interventions. However, existing work has largely focused on analyzing collected data in the off-line setting. In this paper, we propose a novel approach to learning shallow detection cascades developed explicitly for use in a real-time wearable-phone or wearable-phone-cloud
more » ... stems. We apply our approach to the problem of cigarette smoking detection from a combination of wrist-worn actigraphy data and respiration chest band data using two and three stage cascades.
arXiv:1607.03730v1 fatcat:r25rcsbezvdhlpirqre7mbrlty

Masking schemes for image manifolds

Hamid Dadkhahi, Marco F. Duarte
2014 2014 IEEE Workshop on Statistical Signal Processing (SSP)  
We consider the problem of selecting an optimal mask for an image manifold, i.e., choosing a subset of the dimensions of the image space that preserves the manifold structure present in the original data. Such masking implements a form of compressed sensing that reduces power consumption in emerging imaging sensor platforms. Our goal is for the manifold learned from masked images to resemble the manifold learned from full images as closely as possible. We show that the process of finding the
more » ... imal masking pattern can be cast as a binary integer program, which is computationally expensive but can be approximated by a fast greedy algorithm. Numerical experiments show that the manifolds learned from masked images resemble those learned from full images for modest mask sizes. Furthermore, our greedy algorithm performs similarly to the exhaustive search from integer programming at a fraction of the computational cost.
doi:10.1109/ssp.2014.6884623 dblp:conf/ssp/DadkhahiD14 fatcat:diztakrdtrbynbfvuujbone5ay

Design Of Higher Density Dual-Tree Discrete Wavelet Transform With Few Degrees Of Freedom

Hamid Dadkhahi, Bogdan Dumitrescu
2008 Zenodo  
Publication in the conference proceedings of EUSIPCO, Lausanne, Switzerland, 2008
doi:10.5281/zenodo.40888 fatcat:2zl3sn3xxnhrrp2ujdg2zwbjce

Spectral Compressive Sensing with Polar Interpolation [article]

Karsten Fyhn, Hamid Dadkhahi, Marco F. Duarte
2013 arXiv   pre-print
Existing approaches to compressive sensing of frequency-sparse signals focuses on signal recovery rather than spectral estimation. Furthermore, the recovery performance is limited by the coherence of the required sparsity dictionaries and by the discretization of the frequency parameter space. In this paper, we introduce a greedy recovery algorithm that leverages a band-exclusion function and a polar interpolation function to address these two issues in spectral compressive sensing. Our
more » ... m is geared towards line spectral estimation from compressive measurements and outperforms most existing approaches in fidelity and tolerance to noise.
arXiv:1303.2799v3 fatcat:ap6g2t552jgsdlif3vj56fcmda

Combinatorial Black-Box Optimization with Expert Advice [article]

Hamid Dadkhahi, Karthikeyan Shanmugam, Jesus Rios, Payel Das, Samuel Hoffman, Troy David Loeffler, Subramanian Sankaranarayanan
2020 arXiv   pre-print
We consider the problem of black-box function optimization over the boolean hypercube. Despite the vast literature on black-box function optimization over continuous domains, not much attention has been paid to learning models for optimization over combinatorial domains until recently. However, the computational complexity of the recently devised algorithms are prohibitive even for moderate numbers of variables; drawing one sample using the existing algorithms is more expensive than a function
more » ... valuation for many black-box functions of interest. To address this problem, we propose a computationally efficient model learning algorithm based on multilinear polynomials and exponential weight updates. In the proposed algorithm, we alternate between simulated annealing with respect to the current polynomial representation and updating the weights using monomial experts' advice. Numerical experiments on various datasets in both unconstrained and sum-constrained boolean optimization indicate the competitive performance of the proposed algorithm, while improving the computational time up to several orders of magnitude compared to state-of-the-art algorithms in the literature.
arXiv:2006.03963v2 fatcat:gvrzyah5ibdijbjskzbxduqo6e

Inverse polynomial reconstruction method in DCT domain

Hamid Dadkhahi, Atanas Gotchev, Karen Egiazarian
2012 EURASIP Journal on Advances in Signal Processing  
The discrete cosine transform (DCT) offers superior energy compaction properties for a large class of functions and has been employed as a standard tool in many signal and image processing applications. However, it suffers from spurious behavior in the vicinity of edge discontinuities in piecewise smooth signals. To leverage the sparse representation provided by the DCT, in this article, we derive a framework for the inverse polynomial reconstruction in the DCT expansion. It yields the
more » ... of a piecewise smooth signal in terms of polynomial coefficients, obtained from the DCT representation of the same signal. Taking advantage of this framework, we show that it is feasible to recover piecewise smooth signals from a relatively small number of DCT coefficients with high accuracy. Furthermore, automatic methods based on minimum description length principle and cross-validation are devised to select the polynomial orders, as a requirement of the inverse polynomial reconstruction method in practical applications. The developed framework can considerably enhance the performance of the DCT in sparse representation of piecewise smooth signals. Numerical results show that denoising and image approximation algorithms based on the proposed framework indicate significant improvements over wavelet counterparts for this class of signals.
doi:10.1186/1687-6180-2012-133 fatcat:34fxhn4eznel5jcamd6tv4rvci

Fourier Representations for Black-Box Optimization over Categorical Variables [article]

Hamid Dadkhahi, Jesus Rios, Karthikeyan Shanmugam, Payel Das
2022 arXiv   pre-print
The sparsity parameter λ in exponential weight updates is set to 1 in all the experiments following the same choice made in (Dadkhahi et al. 2020) .  ...  Surrogate Model Learning Sparse Online Regression: We adopt the learning algorithm of combinatorial optimization with expert advice (Dadkhahi et al. 2020) in the following way.  ... 
arXiv:2202.03712v2 fatcat:maadiycz7zd23c7oxvpvh3iywa

Image masking schemes for local manifold learning methods

Hamid Dadkhahi, Marco F. Duarte
2015 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
We consider the problem of selecting a subset of the dimensions of an image manifold that best preserves the underlying local structure in the original data. We have previously shown that masks which preserve the data neighborhood graph are well suited to global manifold learning algorithms. However, local manifold learning algorithms leverage a geometric structure beyond that captured by this neighborhood graph. In this paper, we present a mask selection algorithm that further preserves this
more » ... ditional structure by designing an extended data neighborhood graph that connects all neighbors of each data point, forming local cliques. Numerical experiments show the improvements achieved by employing the extended graph in the mask selection process.
doi:10.1109/icassp.2015.7179077 dblp:conf/icassp/DadkhahiD15 fatcat:2j3cu5rlbvat7pluxt3ucq5wh4

Spectral compressive sensing with polar interpolation

Karsten Fyhn, Hamid Dadkhahi, Marco F. Duarte
2013 2013 IEEE International Conference on Acoustics, Speech and Signal Processing  
Existing approaches to compressive sensing of frequencysparse signals focuses on signal recovery rather than spectral estimation. Furthermore, the recovery performance is limited by the coherence of the required sparsity dictionaries and by the discretization of the frequency parameter space. In this paper, we introduce a greedy recovery algorithm that leverages a band-exclusion function and a polar interpolation function to address these two issues in spectral compressive sensing. Our
more » ... is geared towards line spectral estimation from compressive measurements and outperforms most existing approaches in fidelity and tolerance to noise.
doi:10.1109/icassp.2013.6638862 dblp:conf/icassp/FyhnDD13 fatcat:xe6h3gm3qbhlnhbhc4ujef74aa

Out-of-Sample Extension for Dimensionality Reduction of Noisy Time Series

Hamid Dadkhahi, Marco F. Duarte, Benjamin M. Marlin
2017 IEEE Transactions on Image Processing  
Dadkhahi high-dimensional ambient space despite possessing merely a few degrees of freedom.  ... 
doi:10.1109/tip.2017.2735189 pmid:28783635 fatcat:ttkwz4zmvzfqjgwyrkrstqu7bq

Learning Tree-Structured Detection Cascades for Heterogeneous Networks of Embedded Devices

Hamid Dadkhahi, Benjamin M. Marlin
2017 Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17  
Dadkhahi To begin, we assume we have access to a total of D + is defined as shown in Equation 2.  ...  Dadkhahi Figure 5 . 5 In this figure, timing results are averaged over 1, 000 classifier evaluations.  ... 
doi:10.1145/3097983.3098169 pmid:29333328 pmcid:PMC5765542 dblp:conf/kdd/DadkhahiM17 fatcat:fnjpeeb575hsfa2ektwsozdyt4