Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (6 results)

  • Thesis
  • Peer Reviewed
UC San Diego Electronic Theses and Dissertations (2009)
With the increasing areal density in magnetic recording systems, perpendicular recording has replaced longitudinal recording to overcome the superparamagnetic limit. Studies on perpendicular recording channels including aspects of channel modeling, signal processing and coding techniques are presented in this dissertation. To optimize a high density perpendicular magnetic recording system, one needs to know the tradeoffs between various components of the system including the read/write transducers, the magnetic medium, and the read channel. We extend the work by Chaichanavong on the parameter optimization for systems via design curves. Different signal processing and coding techniques are studied. Information-theoretic tools are utilized to determine the acceptable region for the channel parameters when optimal detection and linear coding techniques are used. Our results show that a considerable gain can be achieved by the optimal detection and coding techniques. The read-write process in perpendicular magnetic recording channels includes a number of nonlinear effects. Nonlinear transition shift (NLTS) is one of them. The signal distortion induced by NLTS can be reduced by write precompensation during data recording. We numerically evaluate the effect of NLTS on the read-back signal and examine the effectiveness of several write precompensation schemes in combating NLTS in a channel characterized by both transition jitter noise and additive white Gaussian electronics noise. We also present an analytical method to estimate the bit-error- rate and use it to help determine the optimal write precompensation values in multi-level precompensation schemes. We propose a mean-adjusted pattern-dependent noise predictive (PDNP) detection algorithm for use on the channel with NLTS. We show that this detector can offer significant improvements in bit-error-rate (BER) compared to conventional Viterbi and PDNP detectors. Moreover, the system performance can be further improved by combining the new detector with a simple write precompensation scheme. Soft-decision decoding for algebraic codes can improve performance for magnetic recording systems. In this dissertation, we propose two soft-decision decoding methods for tensor-product parity codes. We also present a list decoding algorithm for generalized error locating codes
    Cover page: Channel modeling, signal processing and coding for perpendicular magnetic recording
    • Thesis
    • Peer Reviewed
    UC Berkeley Electronic Theses and Dissertations (2024)

    Autonomous robots have swiftly revolutionized several industries, enhancing manufacturing processes by streamlining assembly lines for heightened efficiency, revolutionizing agriculture with automated planting and harvesting, and refining logistics through the optimization of warehousing and delivery systems. These advancements underscore the groundbreaking impact of robotics on productivity and innovation across a diverse range of sectors. However, the rapid advancement in autonomous robotics faces a significant challenge: the heavy dependency on massive data and the daunting task of generalizing learned behaviors to new tasks and environments. This challenge represents more than a mere technical obstacle; it is a critical bottleneck that constrains the widespread adoption and effectiveness of autonomous robots, stifling innovation and practical deployment across a myriad of industrial applications.

    In this dissertation, we study the problem of data-efficient learning and generalization for industrial autonomous robots. Our goal is to develop algorithms that enable robots to learn from limited data and generalize the learned behaviors to novel tasks and environments effectively. The core idea is to leverage proper task knowledge and assumptions, embed- ding them into the algorithmic designs to significantly enhance their data efficiency. Our research endeavors are dedicated to three principal aspects: data-efficient reward learning, data-efficient policy learning, and data-efficient policy generalization. These approaches are meticulously applied across a wide array of industrial scenarios, such as autonomous vehicles, robotic assembly, and robotic palletization, showcasing their versatility and effectiveness in enhancing robotic efficiency and adaptability in real-world applications.

    This dissertation unfolds in three distinct parts, offering a comprehensive examination of data-efficient learning and generalization for autonomous robots. Part I lays the foundation with an in-depth exploration of data-efficient reward learning, employing inverse reinforcement learning (Chapter 2) and representation learning (Chapter 3) to uncover efficient ways to infer reward signals from limited data. Part II shifts focus to the nuances of data-efficient policy learning. Chapter 4 introduces a data-efficient reinforcement learning (RL) policy specifically designed for robotic palletization, enhancing data efficiency by narrowing the exploration space via learned action space masking. In Chapter 5, we propose a novel skill representation method, namely motion primitives (MP), alongside a data-efficient framework for learning MP-based insertion skills directly from human demonstrations. Concluding with Part III, the dissertation advances into the realm of data-efficient policy generalization across diverse tasks. In Chapter 6, a novel zero-shot policy generalization approach is presented, capitalizing on the compositional structure of task representations to enable seamless adaptation to new tasks without the need for additional data.

      Cover page: Data-Efficient Learning and Generalization for Industrial Robotic Systems
      • Article
      • Peer Reviewed
      LBL Publications (2019)
      By investigating the bulk superconducting state via dc magnetization measurements, we have discovered a common resurgence of the superconducting transition temperatures (Tcs) of the monolayer Bi2Sr2CuO6+δ (Bi2201) and bilayer Bi2Sr2CaCu2O8+δ (Bi2212) to beyond the maximum Tcs (Tc-maxs) predicted by the universal relation between Tc and doping (p) or pressure (P) at higher pressures. The Tc of underdoped Bi2201 initially increases from 9.6 K at ambient to a peak at 23 K at 26 GPa and then drops as expected from the universal Tc-P relation. However, at pressures above 40 GPa, Tc rises rapidly without any sign of saturation up to 30 K at 51 GPa. Similarly, the Tc for the slightly overdoped Bi2212 increases after passing a broad valley between 20 and 36 GPa and reaches 90 K without any sign of saturation at 56 GPa. We have, therefore, attributed this Tc resurgence to a possible pressure-induced electronic transition in the cuprate compounds due to a charge transfer between the Cu 3[Formula: see text] and the O 2p bands projected from a hybrid bonding state, leading to an increase of the density of states at the Fermi level, in agreement with our density functional theory calculations. Similar Tc-P behavior has also been reported in the trilayer Br2Sr2Ca2Cu3O10+δ (Bi2223). These observations suggest that higher Tcs than those previously reported for the layered cuprate high-temperature superconductors can be achieved by breaking away from the universal Tc-P relation through the application of higher pressures.
        Cover page: Higher superconducting transition temperature by breaking the universal pressure relation
        • Article
        • Peer Reviewed
        LBL Publications (2018)
        Metagenomics, the study of all microbial species cohabitants in an environment, often produces large amount of sequence data varying from several GBs to a few TBs. Analyzing metagenomics data includes both data-intensive and compute-intensive steps, making the entire process hard to scale. Here we aim to optimize a metagenomics application that partitions the shortgun metagenomics sequences based on their species of origin. Our solution combines MapReduce-based BioPig analytic toolkit with MPI to provide scalability in respective to both data and compute. We also made some improvements to the existing BioPig toolkit by using simplified data types and compressed k-mer storage. These optimizations leads up to 193× speedup for the computing-intensive step and 9.6× speedup over the entire pipeline. Our optimized application is also capable of processing datasets that are 16 times larger on the same hardware platform. These results suggest integrating heterogeneous technologies such as Hadoop and MPI is quite efficient to solve large genomics problems that are both data-intensive and compute-intensive.
          Cover page: Combining Hadoop with MPI to Solve Metagenomics Problems that are both Data- and Compute-intensive
          • Article
          • Peer Reviewed
          UC San Diego Previously Published Works (2016)
          Rett syndrome is a severe form of autism spectrum disorder, mainly caused by mutations of a single gene methyl CpG binding protein 2 (MeCP2) on the X chromosome. Patients with Rett syndrome exhibit a period of normal development followed by regression of brain function and the emergence of autistic behaviors. However, the mechanism behind the delayed onset of symptoms is largely unknown. Here we demonstrate that neuron-specific K(+)-Cl(-) cotransporter2 (KCC2) is a critical downstream gene target of MeCP2. We found that human neurons differentiated from induced pluripotent stem cells from patients with Rett syndrome showed a significant deficit in KCC2 expression and consequently a delayed GABA functional switch from excitation to inhibition. Interestingly, overexpression of KCC2 in MeCP2-deficient neurons rescued GABA functional deficits, suggesting an important role of KCC2 in Rett syndrome. We further identified that RE1-silencing transcriptional factor, REST, a neuronal gene repressor, mediates the MeCP2 regulation of KCC2. Because KCC2 is a slow onset molecule with expression level reaching maximum later in development, the functional deficit of KCC2 may offer an explanation for the delayed onset of Rett symptoms. Our studies suggest that restoring KCC2 function in Rett neurons may lead to a potential treatment for Rett syndrome.
            Cover page: KCC2 rescues functional deficits in human neurons derived from patients with Rett syndrome
            • Article
            • Peer Reviewed
            UC Irvine Previously Published Works (2024)
            Rosacea is a chronic inflammatory skin disorder, whose underlying cellular and molecular mechanisms remain obscure. Here, we generate a single-cell atlas of facial skin from female rosacea patients and healthy individuals. Among keratinocytes, a subpopulation characterized by IFNγ-mediated barrier function damage is found to be unique to rosacea lesions. Blocking IFNγ signaling alleviates rosacea-like phenotypes and skin barrier damage in mice. The papulopustular rosacea is featured by expansion of pro-inflammatory fibroblasts, Schwann, endothelial and macrophage/dendritic cells. The frequencies of type 1/17 and skin-resident memory T cells are increased, and vascular mural cells are characterized by activation of inflammatory pathways and impaired muscle contraction function in rosacea. Most importantly, fibroblasts are identified as the leading cell type producing pro-inflammatory and vasodilative signals in rosacea. Depletion of fibroblasts or knockdown of PTGDS, a gene specifically upregulated in fibroblasts, blocks rosacea development in mice. Our study provides a comprehensive understanding of the aberrant alterations of skin-resident cell populations and identifies fibroblasts as a key determinant in rosacea development.
              Cover page: Single-cell transcriptomics reveals aberrant skin-resident cell populations and identifies fibroblasts as a determinant in rosacea.
              Top