A5017364251- Cancer Genomics and Diagnostics
- Advanced Adaptive Filtering Techniques
- Pancreatic and Hepatic Oncology Research
- Neural Networks and Applications
- Blind Source Separation Techniques
- Single-cell and spatial transcriptomics
- Human Mobility and Location-Based Analysis
- Advanced Optimization Algorithms Research
- Epigenetics and DNA Methylation
- Transportation Planning and Optimization
- Economic theories and models
- Optimization and Variational Analysis
- Game Theory and Applications
- Matrix Theory and Algorithms
- Data Management and Algorithms
- Cryptography and Residue Arithmetic
- Gene expression and cancer classification
- Advanced Thermodynamics and Statistical Mechanics
- Auction Theory and Applications
- Numerical Methods and Algorithms
- Polynomial and algebraic computation
- Supply Chain and Inventory Management
- Network Packet Processing and Optimization
- Distributed Control Multi-Agent Systems
- Transportation and Mobility Innovations
University of California, Berkeley
2021-2024
Center for Information Technology Research in the Interest of Society
2024
Lawrence Berkeley National Laboratory
2020-2023
University of California, Santa Barbara
2015-2016
Bilkent University
2016
Koç University
2004-2005
The software and data in this repository are a snapshot of the that were used research reported on paper Sempervirens: A Fast Reconstruction Algorithm for Noisy Incomplete Binary Matrix Representations Trees. is based SHA development repository.
Applications such as reconstructing cell lineage trees (represented phylogenetic trees) from single-cell sequencing data require a [Formula: see text]-matrix that has many errors and missing entries. We introduce Sempervirens, very fast matrix reconstruction algorithm for noisy incomplete representations of trees. Sempervirens uses an iterative maximum-likelihood approach to determine the topology tree represented by corrupted data. show is at least three orders magnitude faster than other...
We propose a novel blind equalization method based on subgradient search over convex cost surface. This is an alternative to the existing iterative approaches such as Constant Modulus Algorithm (CMA), which often suffer from convergence problems caused by their nonconvex functions. The proposed algorithm called SubGradient Blind (SGBA) for both real and complex constellations, with very simple update rule. It minimization of l/sub /spl infin// norm equalizer output under linear constraint...
Abstract Advances in single-cell RNA sequencing (scRNAseq) technologies uncovered an unexpected complexity tumors, underlining the relevance of intratumor heterogeneity to cancer progression and therapeutic resistance. Heterogeneity mutational composition cells is a result distinct (sub)clonal expansions, each with metastatic potential resistance specific treatments. Unfortunately, due their low read coverage per cell, scRNAseq datasets are too sparse noisy be used for detecting expressed...
We introduce a matrix decomposition method and prove that multiplication in GF <inline-formula> <tex-math notation="LaTeX">$(2^k)$</tex-math></inline-formula> with Type 1 optimal normal basis for can be performed using <inline-formula><tex-math notation="LaTeX">$k^2-1$</tex-math></inline-formula> XOR gates irrespective of the choice irreducible polynomial generating field. The previous results achieved this bound only special polynomials. Furthermore, performs operation...
We leverage best response dynamics to solve monotone variational inequalities on compact and convex sets. Specialization of the method in game theory recovers convergence results Nash equilibria when agents select current distribution strategies. apply generalize population games with additional constraints. Furthermore, we explore robustness by introducing various types time-varying disturbances.
A novel blind equalization method, based on a subgradient search over convex cost surface, is proposed. This an alternative to the existing iterative approaches such as constant modulus algorithm (CMA) which mostly suffer from convergence problems caused by their non-convex functions. The proposed method algorithm, for both real and complex constellations, with very simple update rule that minimizes l/sub /spl infin// norm of equalizer output under linear constraint coefficients. has nice...
Trades based on bilateral (indivisible) contracts can be represented by a network. Vertices correspond to agents, whereas arcs represent the nonprice elements of contract. Given prices for each arc, agents choose incident that maximize their utility. We enlarge model allow polymatroidal constraints set may traded, which interpreted as modeling limited one-for-one substitution. show that, two-sided markets, there exists competitive equilibrium; however, multisided this not possible. Funding:...
A novel blind equalization method based on a subgradient search over convex cost surface is examined under noisy channel and modification proposed. This an alternative to the existing iterative approaches such as constant modulus algorithm (CMA) which mostly suffer from convergence problems caused by their non-convex functions. The proposed algorithm, for both real complex constellations, with very simple update rule that minimizes l/sub /spl infin// norm of equalizer output linear...
Trades based on bilateral (indivisible) contracts can be represented by a network. Vertices correspond to agents while arcs represent the non-price elements of contract. Given prices for each arc, choose incident that maximize their utility. We enlarge model allow polymatroidal constraints set may traded which interpreted as modeling limited one for-one substitution. show two-sided markets there exists competitive equilibrium however multi-sided this not possible.
Abstract Emerging sets of single-cell sequencing data makes it appealing to apply existing tumor phylogeny reconstruction methods analyze associated intratumor heterogeneity. Unfortunately, inference is an NP-hard problem and principled typically fail scale up handle thousands cells mutations observed in emerging sets. Even though there are greedy heuristics build hierarchical clustering mutations, they suffer from well-documented issues accuracy. Additionally even when “optimal” solutions...
The Braess paradox is a counter-intuitive phenomenon whereby adding roads to network results in higher travel time at equilibrium. In this paper we present an algorithm detect the occurrence of real-world networks with help improved graph representation accounting for queues. addition queues enables closer match real data. Moreover, search routes causing ("Braess routes") rather than links, and advocate removing such virtually from navigation systems so that associated links can continue...