The Experts below are selected from a list of 306 Experts worldwide ranked by ideXlab platform

Alp Akcay - One of the best experts on this subject based on the ideXlab platform.

  • Reinforcement Learning under Model Risk for Biomanufacturing Fermentation Control.
    arXiv: Machine Learning, 2021
    Co-Authors: Bo Wang, Wei Xie, Tugce Martagan, Alp Akcay
    Abstract:

    In the biopharmaceutical manufacturing, Fermentation process plays a critical role impacting on productivity and profit. Since biotherapeutics are manufactured in living cells whose biological mechanisms are complex and have highly variable outputs, in this paper, we introduce a model-based reinforcement learning framework accounting for model risk to support bioprocess online learning and guide the optimal and robust customized stopping policy for Fermentation process. Specifically, built on the dynamic mechanisms of protein and impurity generation, we first construct a probabilistic model characterizing the impact of underlying bioprocess stochastic uncertainty on impurity and protein growth rates. Since biopharmaceutical manufacturing often has very limited data during the development and early stage of production, we derive the posterior distribution quantifying the process model risk, and further develop the Bayesian rule based knowledge update to support the online learning on underlying stochastic process. With the prediction risk accounting for both bioprocess stochastic uncertainty and model risk, the proposed reinforcement learning framework can proactively hedge all sources of uncertainties and support the optimal and robust customized decision making. We conduct the structural analysis of optimal policy and study the impact of model risk on the policy selection. We can show that it asymptotically converges to the optimal policy obtained under perfect information of underlying stochastic process. Our case studies demonstrate that the proposed framework can greatly improve the biomanufacturing industrial practice.

Bo Wang - One of the best experts on this subject based on the ideXlab platform.

  • Reinforcement Learning under Model Risk for Biomanufacturing Fermentation Control.
    arXiv: Machine Learning, 2021
    Co-Authors: Bo Wang, Wei Xie, Tugce Martagan, Alp Akcay
    Abstract:

    In the biopharmaceutical manufacturing, Fermentation process plays a critical role impacting on productivity and profit. Since biotherapeutics are manufactured in living cells whose biological mechanisms are complex and have highly variable outputs, in this paper, we introduce a model-based reinforcement learning framework accounting for model risk to support bioprocess online learning and guide the optimal and robust customized stopping policy for Fermentation process. Specifically, built on the dynamic mechanisms of protein and impurity generation, we first construct a probabilistic model characterizing the impact of underlying bioprocess stochastic uncertainty on impurity and protein growth rates. Since biopharmaceutical manufacturing often has very limited data during the development and early stage of production, we derive the posterior distribution quantifying the process model risk, and further develop the Bayesian rule based knowledge update to support the online learning on underlying stochastic process. With the prediction risk accounting for both bioprocess stochastic uncertainty and model risk, the proposed reinforcement learning framework can proactively hedge all sources of uncertainties and support the optimal and robust customized decision making. We conduct the structural analysis of optimal policy and study the impact of model risk on the policy selection. We can show that it asymptotically converges to the optimal policy obtained under perfect information of underlying stochastic process. Our case studies demonstrate that the proposed framework can greatly improve the biomanufacturing industrial practice.

  • Reinforcement Learning under Model Risk for Biomanufacturing Fermentation Control
    2021
    Co-Authors: Bo Wang, Xie Wei, Martagan Tugce, Akcay Alp
    Abstract:

    In the biopharmaceutical manufacturing, Fermentation process plays a critical role impacting on productivity and profit. Since biotherapeutics are manufactured in living cells whose biological mechanisms are complex and have highly variable outputs, in this paper, we introduce a model-based reinforcement learning framework accounting for model risk to support bioprocess online learning and guide the optimal and robust customized stopping policy for Fermentation process. Specifically, built on the dynamic mechanisms of protein and impurity generation, we first construct a probabilistic model characterizing the impact of underlying bioprocess stochastic uncertainty on impurity and protein growth rates. Since biopharmaceutical manufacturing often has very limited data during the development and early stage of production, we derive the posterior distribution quantifying the process model risk, and further develop the Bayesian rule based knowledge update to support the online learning on underlying stochastic process. With the prediction risk accounting for both bioprocess stochastic uncertainty and model risk, the proposed reinforcement learning framework can proactively hedge all sources of uncertainties and support the optimal and robust customized decision making. We conduct the structural analysis of optimal policy and study the impact of model risk on the policy selection. We can show that it asymptotically converges to the optimal policy obtained under perfect information of underlying stochastic process. Our case studies demonstrate that the proposed framework can greatly improve the biomanufacturing industrial practice.Comment: 37 pages, 7 figure

Jean-marie Sablayrolles - One of the best experts on this subject based on the ideXlab platform.

  • Comprehensive Study of the Evolution of the Gas–Liquid Partitioning of Acetaldehyde during Wine Alcoholic Fermentation
    Journal of Agricultural and Food Chemistry, 2018
    Co-Authors: Evelyne Aguera, Jean-marie Sablayrolles, Yannick Sire, Jean-roch Mouret, Vincent Farines
    Abstract:

    Determining the gas-liquid partitioning (K-i) of acetaldehyde during alcoholic Fermentation is an important step in the optimization of Fermentation Control with the aim of minimizing the accumulation of this compound, which is responsible for the undesired attributes of green apples and fresh-cut grass in wines. In this work, the effects of the main Fermentation parameters on the K-i of acetaldehyde were assessed. K-i values were found to be dependent on the temperature and composition of the medium. A nonlinear correlation between the evolution of the K-i and Fermentation progress was observed, attributable to the strong retention effect of ethanol at low concentrations, and it was demonstrated that the partitioning of this specific molecule was not influenced by the CO2 production rate. A model was developed that quantifies the K-i of acetaldehyde with a very accurate prediction, as the difference between the observed and predicted values did not exceed 9%.

  • Review: Characterization and Role of Grape Solids during Alcoholic Fermentation under Enological Conditions
    American Journal of Enology and Viticulture, 2016
    Co-Authors: Erick Casalta, Aude Vernhet, Jean-marie Sablayrolles, Catherine Tesniere, Jean-michel Salmon
    Abstract:

    During wine production, grape solids have a large impact on Fermentation characteristics and the organoleptic qualities of the resulting wine. We review here the research carried out on grape solids. We begin by focusing on the origin, physical characteristics and composition of these solids, and the changes in these aspects occurring during Fermentation. We then consider the effect of solids on Fermentation, the role of sterols, the Control of solids and interactions between solids and other nutrients. Solids exert their effects on alcoholic Fermentation mainly by modulating lipid supply. The balance between solid content and nitrogen is a key factor in Fermentation Control. The study of grape solids is recent and requires further development. Knowledge of the composition of these solids, and of sterol uptake mechanisms by yeast should facilitate improvements in Fermentation Control.

  • Review: Characterization and Role of Grape Solids during Alcoholic Fermentation under Enological Conditions
    American Journal of Enology and Viticulture, 2015
    Co-Authors: Erick Casalta, Aude Vernhet, Jean-marie Sablayrolles, Catherine Tesniere, Jean-michel Salmon
    Abstract:

    During wine production, grape solids have a large impact on the Fermentation characteristics and organoleptic qualities of the resulting wine. Here we review the research on grape solids. We begin by focusing on the origin, physical characteristics, and composition of these solids and on the changes in these factors that occur during Fermentation. We then consider the impact of solids on Fermentation, the role of sterols, the Control of solids, and interactions between solids and other nutrients. Solids exert their effects on alcoholic Fermentation mainly by modulating lipid supply. The balance between solids content and nitrogen is a key factor in Fermentation Control. The study of grape solids is in its infancy and requires further development. Knowledge of the composition of these solids and of sterol uptake mechanisms by yeast should facilitate improvements in Fermentation Control.

Mariefrancoise Gorwagrauslund - One of the best experts on this subject based on the ideXlab platform.

  • increased tolerance and conversion of inhibitors in lignocellulosic hydrolysates by saccharomyces cerevisiae
    Journal of Chemical Technology & Biotechnology, 2007
    Co-Authors: Joao R M Almeida, Tobias Modig, Anneli Petersson, Barbel Hahnhagerdal, Gunnar Liden, Mariefrancoise Gorwagrauslund
    Abstract:

    During hydrolysis of lignocellulosic biomass, monomeric sugars and a broad range of inhibitory compounds are formed and released. These inhibitors, which can be organized around three main groups, furans, weak acids and phenolics, reduce ethanol yield and productivity by affecting the microorganism performance during the Fermentation step. Among the microorganisms that have been evaluated for lignocellulosic hydrolysate ethanol Fermentation, the yeast Saccharomyces cerevisiae appears to be the least sensitive. In order to overcome the effect of inhibitors, strategies that include improvement of natural tolerance of microorganism and use of Fermentation Control strategies have been developed. An overview of the origin, effects and mechanisms of action of known inhibitors on S. cerevisiae is given. Fermentation Control strategies as well as metabolic, genetic and evolutionary engineering strategies to obtain S. cerevisiae strains with improved tolerance are discussed.

He Peng - One of the best experts on this subject based on the ideXlab platform.

  • Design of Biology Fermentation Control System Based on Fuzzy Neural Network and PSO
    Computer Simulation, 2012
    Co-Authors: He Peng
    Abstract:

    Biological Fermentation process is of high variability and nonlinear,therefore,it is difficult to establish the precise mathematical model of the system.Using particle swarm optimization algorithm and fuzzy neural network with their respective advantages,we put forward a particle swarm algorithm of fuzzy neural network,and applied biology Fermentation Control system.By using fuzzy neural network to build biological Fermentation Control system,the particle swarm optimization algorithm for fuzzy neural network parameters was optimized.Finally,the simulation experiment was carried out to test the Control system performances.The simulation results show that,the fuzzy neural network based on particle swarm biology Fermentation Control system has higher Control precision,stronger robustness,and has good application prospect.