1,721,068 research outputs found
Decision Making in Star Networks with Incorrect Beliefs
No full textConsider a Bayesian binary decision-making problem in star networks, where local agents make selfish decisions independently, and a fusion agent makes a final decision based on aggregated decisions and its own private signal. In particular, we assume all agents have private beliefs for the true prior probability, based on which they perform Bayesian decision making. We focus on the Bayes risk of the fusion agent and counterintuitively find that incorrect beliefs could achieve a smaller risk than that when agents know the true prior. It is of independent interest for sociotechnical system design that the optimal beliefs of local agents resemble human probability reweighting models from cumulative prospect theory. We also consider asymptotic characterization of the optimal beliefs and fusion agent's risk in the number of local agents. We find that the optimal risk of the fusion agent converges to zero exponentially fast as the number of local agents grows. Furthermore, having an identical constant belief is asymptotically optimal in the sense of the risk exponent. For additive Gaussian noise, the optimal belief turns out to be a simple function of only error costs and the risk exponent can be explicitly characterized. © 2021 IEEE1
Distributed scalar quantization for computing: High-resolution analysis and extensions
Get PDFCommunication of quantized information is frequently followed by a computation. We consider situations of distributed functional scalar quantization: distributed scalar quantization of (possibly correlated) sources followed by centralized computation of a function. Under smoothness conditions on the sources and function, companding scalar quantizer designs are developed to minimize mean-squared error (MSE) of the computed function as the quantizer resolution is allowed to grow. Striking improvements over quantizers designed without consideration of the function are possible and are larger in the entropy-constrained setting than in the fixed-rate setting. As extensions to the basic analysis, we characterize a large class of functions for which regular quantization suffices, consider certain functions for which asymptotic optimality is achieved without arbitrarily fine quantization, and allow limited collaboration between source encoders. In the entropy-constrained setting, a single bit per sample communicated between encoders can have an arbitrarily large effect on functional distortion. In contrast, such communication has very little effect in the fixed-rate setting.National Science Foundation (U.S.) (Grant 0729069
Missing values imputation and image registration for genetics applications
Get PDFIn this thesis, we address several common scenarios of corrupted data in data and image processing pipelines. The first is in the setting of clustered data with missing values. We design an algorithm for imputing missing values using optimal recovery and derive an error bound for non-negative matrix factorization of the imputed data. Second, we consider missing values as erasure channels and show examples of using Fano's inequality to find lower bounds on missing values algorithms. Finally, we perform image registration of misaligned and noisy images using multiinformation and use fi nite rate of innovation sample to speed up registration while preserving optimality.Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2019-08-22 without embargo termsThe student, Rebecca Chen, accepted the attached license on 2019-04-24 at 15:11.The student, Rebecca Chen, submitted this Thesis for approval on 2019-04-24 at 15:15.This Thesis was approved for publication on 2019-04-24 at 16:57.DSpace SAF Submission Ingestion Package generated from Vireo submission #13888 on 2019-08-22 at 14:46:38Made available in DSpace on 2019-08-23T20:02:09Z (GMT). No. of bitstreams: 2
CHEN-THESIS-2019.pdf: 2981990 bytes, checksum: 3753f0a43464e667d606cd360c0126f8 (MD5)
LICENSE.txt: 4209 bytes, checksum: 1cd339952afe715b79dbd61cc5e1e3a9 (MD5)
Previous issue date: 2019-04-2
Science of science: Biological research network
Get PDFThe science of science studies how scientists do research effectively. In this thesis, we focus on the science of biology by analyzing a biological research network which is defined as a graph where species are nodes and common papers between species are links. Building upon the idea that many biological papers relate and compare at least two species, we evaluate a series of hypotheses using various analysis techniques and find the importance of model organisms and humans in biological research. Based on these findings, we analyze whether these species are treated differently in biological research. On a global scale, we try to predict sleeping beauty species that may become important in the future despite not being popular at present and try to predict what pairings of species may be influential in biology.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-05-01The student, Ruby Zhuang, accepted the attached license on 2019-04-24 at 20:22.The student, Ruby Zhuang, submitted this Thesis for approval on 2019-04-24 at 20:23.This Thesis was approved for publication on 2019-04-25 at 08:40.DSpace SAF Submission Ingestion Package generated from Vireo submission #13887 on 2019-08-22 at 15:08:31Made available in DSpace on 2019-08-23T20:36:10Z (GMT). No. of bitstreams: 2
ZHUANG-THESIS-2019.pdf: 2193512 bytes, checksum: 2373b18d678fdef8dbb677ab267aa749 (MD5)
LICENSE.txt: 4208 bytes, checksum: 5b9977b366600bc05504dcbbd34cf626 (MD5)
Previous issue date: 2019-04-25Embargo set by: Seth Robbins for item 112208
Lift date: 2021-08-23T20:36:18Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112208 on 2021-08-24T09:15:38Z
Frame permutation quantization
Get PDFFrame permutation quantization (FPQ) is a new vector quantization technique using finite frames. In FPQ, a vector is encoded using a permutation source code to quantize its frame expansion. This means that the encoding is a partial ordering of the frame expansion coefficients. Compared to ordinary permutation source coding, FPQ produces a greater number of possible quantization rates and a higher maximum rate. Various representations for the partitions induced by FPQ are presented and reconstruction algorithms based on linear programming and quadratic programming are derived. Reconstruction using the canonical dual frame is also studied, and several results relate properties of the analysis frame to whether linear reconstruction techniques provide consistent reconstructions. Simulations for Gaussian sources show performance improvements over entropy-constrained scalar quantization for certain combinations of vector dimension and coding rate.National Science Foundation (U.S.) (NSF grant 0729069
A study on creativity: Detection and network structures
Get PDFIn recent years, the topic of creativity has attracted extensive focus in the form of public discussion as well as research study. This has largely been in two areas: applying technology in order to innovate, as well as studying creativity in society and analyzing its dependence on social parameters and on network characteristics. Computational creativity has been used positively for automated creation of new content like art or recipes, but it is also being applied for pernicious activities like generating vile or misleading content, morphing pornographic or unethical videos/pictures to spread misinformation, or for blackmail. Such instances of fake content generated by artificial intelligence based generative techniques with potentially harmful applications are commonly referred to as deepfakes. This thesis consists of two parts that focus on each of these aspects separately. The first part deals with the detection problem for deepfake content. It outlines a classification problem for identifying an image as legitimate or fake, and obtains bounds on the expected performance while identifying fake content generated by generative adversarial networks. It further uses an approximation from Euclidean information theory for the low error regime and gives simplified bounds for the case where accuracy of the generative process is high. The second part deals with studying the effects of network parameters on creative productivity in social networks. It includes an overview of various theories on the ways by which network structure affects creativity, along with empirical results obtained by analyzing university innovation data alongside the online friendship networks for the same universities.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-05-01The student, Sakshi Agarwal, accepted the attached license on 2019-04-24 at 14:03.The student, Sakshi Agarwal, submitted this Thesis for approval on 2019-04-24 at 14:08.This Thesis was approved for publication on 2019-04-24 at 14:53.DSpace SAF Submission Ingestion Package generated from Vireo submission #13870 on 2019-08-22 at 15:08:01Made available in DSpace on 2019-08-23T20:36:08Z (GMT). No. of bitstreams: 3
AGARWAL-THESIS-2019.pdf: 4026043 bytes, checksum: 9735efe1c8604ff24e151ca80aeccb34 (MD5)
Sakshi-MS Thesis.zip: 3115644 bytes, checksum: 5750ff97e9be7aaf22378e4385712998 (MD5)
LICENSE.txt: 4211 bytes, checksum: 06c4036f5ef1db318824e02bdf7637b4 (MD5)
Previous issue date: 2019-04-24Embargo set by: Seth Robbins for item 112202
Lift date: 2021-08-23T20:36:18Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112202 on 2021-08-24T09:15:35Z
Structural Properties of the Caenorhabditis elegans Neuronal Network
Get PDFDespite recent interest in reconstructing neuronal networks, complete wiring diagrams on the level of individual synapses remain scarce and the insights into function they can provide remain unclear. Even for Caenorhabditis elegans, whose neuronal network is relatively small and stereotypical from animal to animal, published wiring diagrams are neither accurate nor complete and self-consistent. Using materials from White et al. and new electron micrographs we assemble whole, self-consistent gap junction and chemical synapse networks of hermaphrodite C. elegans. We propose a method to visualize the wiring diagram, which reflects network signal flow. We calculate statistical and topological properties of the network, such as degree distributions, synaptic multiplicities, and small-world properties, that help in understanding network signal propagation. We identify neurons that may play central roles in information processing, and network motifs that could serve as functional modules of the network. We explore propagation of neuronal activity in response to sensory or artificial stimulation using linear systems theory and find several activity patterns that could serve as substrates of previously described behaviors. Finally, we analyze the interaction between the gap junction and the chemical synapse networks. Since several statistical properties of the C. elegans network, such as multiplicity and motif distributions are similar to those found in mammalian neocortex, they likely point to general principles of neuronal networks. The wiring diagram reported here can help in understanding the mechanistic basis of behavior by generating predictions about future experiments involving genetic perturbations, laser ablations, or monitoring propagation of neuronal activity in response to stimulation.National Science Foundation (U.S.) (Grant No. 0325774)National Science Foundation (U.S.) (Grant No. 0836720)National Science Foundation (U.S.) (Grant No. 0729069)National Institute of Mental Health (U.S.) (Grant 69838)Swartz FoundationKlingenstein Foundatio
Malleable coding with edit-distance cost
No full textA malleable coding scheme considers not only representation length but also ease of representation update, thereby encouraging some form of recycling to convert an old codeword into a new one. We examine the trade-off between compression efficiency and malleability cost, measured with a string edit distance that introduces a metric topology to the representation domain. We characterize the achievable rates and malleability as the solution of a subgraph isomorphism problem.National Science Foundation (U.S.) (Graduate Research Fellowship)National Science Foundation (U.S.) (CCR-0325774)National Science Foundation (U.S.) (CCR-0325774
Herb & Spice Network and Health Indications
No full textSpices and herbs are essential culinary ingredients used in cuisines all over the world. They are
known to have medicinal values for large varieties of diseases. In this study, we explored the
relationship between cuisines, spices and diseases through a network analysis viewpoint. We
started with constructing an extensive dictionary between medicinal spices and herbs, and their
disease associations text mined from two handbooks. Centrality measures and various clustering
algorithms were applied to the resulting bipartite and projection graphs to identify the spices and
herbs that play the main roles in disease curing and spice groups that share similar therapeutic
values. Minimum set cover problem was established to find the minimum set of spices needed to
cover the target group of diseases. Then we specifically studied the spice usage patterns in Indian
cuisines, based on the recipes collected from the two main Indian culinary websites. The variations
of spice usages across different regions were learned. We further modeled the evolution of regional
cuisines by generating random recipes with copy-mutate algorithms and compared their disease
coverage capabilities with the real recipe data.U of I Only Restriction set for Item 113177 on 2020-01-08T16:35:51Z with date by [email protected] by Janice Progen ([email protected]) on 2020-01-08T16:47:36Z
No. of bitstreams: 1
ECE 499-FA2019-S.Zhang.pdf: 1642096 bytes, checksum: 0d82d4be28e779145ae70eb58870cada (MD5)U of I Only Restriction set for Item 113177 on 2020-01-09T21:19:00Z with date by [email protected] for entry into archive by James Hutchinson ([email protected]) on 2020-01-09T21:19:46Z (GMT) No. of bitstreams: 1
ECE 499-FA2019-S.Zhang.pdf: 1642096 bytes, checksum: 0d82d4be28e779145ae70eb58870cada (MD5)Made available in DSpace on 2020-01-09T21:19:46Z (GMT). No. of bitstreams: 1
ECE 499-FA2019-S.Zhang.pdf: 1642096 bytes, checksum: 0d82d4be28e779145ae70eb58870cada (MD5)
Previous issue date: 2019-12Embargo set by: James Hutchinson for item 113177
Lift date: 10000-01-01
Reason: Undergraduate senior thesis not recommended for open accessUndergraduate senior thesis not recommended for open accessU of I Onl
Herb & Spice Network and Health Indications
No full textSpices and herbs are essential culinary ingredients used in cuisines all over the world. They are
known to have medicinal values for large varieties of diseases. In this study, we explored the
relationship between cuisines, spices and diseases through a network analysis viewpoint. We
started with constructing an extensive dictionary between medicinal spices and herbs, and their
disease associations text mined from two handbooks. Centrality measures and various clustering
algorithms were applied to the resulting bipartite and projection graphs to identify the spices and
herbs that play the main roles in disease curing and spice groups that share similar therapeutic
values. Minimum set cover problem was established to find the minimum set of spices needed to
cover the target group of diseases. Then we specifically studied the spice usage patterns in Indian
cuisines, based on the recipes collected from the two main Indian culinary websites. The variations
of spice usages across different regions were learned. We further modeled the evolution of regional
cuisines by generating random recipes with copy-mutate algorithms and compared their disease
coverage capabilities with the real recipe data.U of I OnlyUndergraduate senior thesis not recommended for open acces
- …
