Active learning from process data

Raju, Gokaraju K.; Cooney, Charles L.

doi:10.1002/aic.690441009

Cited by 20 publications

(13 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of the factors that may vary significantly beyond those in the clinical trial are identified as sources of epistemic uncertainty and are listed in Table . Uncertainty may reduce over time as knowledge is gained from real‐world experiences or increase for periods of time (e.g., when unexpected new knowledge is gained about a benefit or risk). For these reasons, benefit‐risk‐uncertainty management should be ongoing throughout the product lifecycle .…”

Section: Benefit‐risk Analysismentioning

confidence: 99%

Benefit‐Risk Analysis for Decision‐Making: An Approach

Raju

Gurumurthi

Domike

2016

Clin Pharma and Therapeutics

View full text Add to dashboard Cite

The analysis of benefit and risk is an important aspect of decision-making throughout the drug lifecycle. In this work, the use of a benefit-risk analysis approach to support decision-making was explored. The proposed approach builds on the qualitative US Food and Drug Administration (FDA) approach to include a more explicit analysis based on international standards and guidance that enables aggregation and comparison of benefit and risk on a common basis and a lifecycle focus. The approach is demonstrated on six decisions over the lifecycle (e.g., accelerated approval, withdrawal, and traditional approval) using two case studies: natalizumab for multiple sclerosis (MS) and bedaquiline for multidrug-resistant tuberculosis (MDR-TB).

show abstract

Section: Benefit‐risk Analysismentioning

confidence: 99%

Benefit‐Risk Analysis for Decision‐Making: An Approach

Raju

Gurumurthi

Domike

2016

Clin Pharma and Therapeutics

View full text Add to dashboard Cite

show abstract

“…Cohn14 and Cohn et al15 generalized these criteria for other types of neural network training methods, as well as mixtures of Gaussian and generalized linear models. Similar methods have been applied to process development,16 however, none of these approaches address the necessity of quickly determining system maxima or minima in a process development setting. Other similar approaches have been proposed to design experiments using neural network models,17, 18 although all of these approaches require evaluation of complex objective functions, and none allow for the inclusion of information from repeated experiments.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear experimental design using Bayesian regularized neural networks

Block

2007

AIChE Journal

View full text Add to dashboard Cite

in Wiley InterScience (www.interscience.wiley.com).Novel criteria for designing experiments for nonlinear processes are presented. These criteria improve on a previous methodology in that they can be used to suggest a batch of new experiments to perform (as opposed to a single new experiment) and are also optimized for discovering improved optima of the system response. This is accomplished by using information theoretic criterion, which also heuristically penalize experiments that are likely to result in low (nonoptimal) results. While the methods may be applied to any type of nonlinear-nonparametric model (radial basis functions and generalized linear regression), they are here exclusively considered in conjunction with Bayesian regularized feedforward neural networks. A focus on the application of rapid process development, and how to use repeated experiments to optimize the training procedures of Bayesian regularized neural networks is shown. The presented methods are applied to three case studies. The first two case studies involve simulations of one and two-dimensional (2-D) nonlinear regression problems. The third case study involves real historical data from bench-scale fermentations generated in our laboratory. It is shown that using the presented criteria to design new experiments can greatly increase a feedforward neural network's ability to predict global optima.

show abstract

“…(15,(22)(23)(24)(25)(26) An effective teaching technique should engage students actively, stimulate sense of enquiry, and facilitate collaborative learning, through, for example, group work. (22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32) In group-work activity, two or three students can apply a newly learned concept in a short application, such as problem solving, which promotes problem-based learning. (22,25,29,30) Group-design projects, in-class presentations, computer simulations, experiments, would be part of the active learning and deep learning.…”

Section: Effective Teaching and Active Learningmentioning

confidence: 99%

“…(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32) In group-work activity, two or three students can apply a newly learned concept in a short application, such as problem solving, which promotes problem-based learning. (22,25,29,30) Group-design projects, in-class presentations, computer simulations, experiments, would be part of the active learning and deep learning. (28)(29)(30)(31)(32)(33) This would enhance the skill of transferring knowledge in higher order within a course or across courses.…”

Section: Effective Teaching and Active Learningmentioning

confidence: 99%

“…(22,25,29,30) Group-design projects, in-class presentations, computer simulations, experiments, would be part of the active learning and deep learning. (28)(29)(30)(31)(32)(33) This would enhance the skill of transferring knowledge in higher order within a course or across courses. (15,16) Some current educational systems teach students to solve problems using cookbook procedures rather than teaching students how to solve problems in engineering analysis, and a survey 14 shows that in some institutions, both instructors and students believe that there is no urgent need for changing the present educational practices, mainly because of misleading assessment practices.…”

Section: Effective Teaching and Active Learningmentioning

confidence: 99%

See 1 more Smart Citation

Effective Teaching And Active Learning Of Engineering Courses With Workbook Strategy

Demi̇rel

2004 Annual Conference Proceedings

View full text Add to dashboard Cite

Often mismatches between learning and teaching styles arise because students are in majority visual and sensing learners, and most instructors are intuitive and reflective learners. Beside that, textbooks also have their own styles, and their contents, depth of coverage of materials, and organization may affect the teaching and learning environment. Instructor, as the primary selector of the textbook, has the responsibility in providing students with effective teaching strategy. Here, we present a new strategy called the 'workbook strategy," which integrates these four elements: (i) classroom analysis, (ii) use of workbook beside textbook, (iii) group work, and (iv) use of 'blackboard' as information technology. The workbook strategy provides verbal and visual elements of the course material in an organized way, and relates fundamentals to applications. Such strategy may reduce the mismatches between learning and teaching styles, and hence improves active learning, critical thinking, and problem solving skills. Most of the students who are exposed to the workbook strategy have found it very effective in their learning.

show abstract

Active learning from process data

Cited by 20 publications

References 46 publications

Benefit‐Risk Analysis for Decision‐Making: An Approach

Benefit‐Risk Analysis for Decision‐Making: An Approach

Nonlinear experimental design using Bayesian regularized neural networks

Effective Teaching And Active Learning Of Engineering Courses With Workbook Strategy

Contact Info

Product

Resources

About