Robust optimization through neuroevolution.
Ontology highlight
ABSTRACT:
SUBMITTER: Pagliuca P
PROVIDER: S-EPMC6396973 | biostudies-literature | 2019
REPOSITORIES: biostudies-literature
Publications
We propose a method for evolving neural network controllers robust with respect to variations of the environmental conditions (i.e. that can operate effectively in new conditions immediately, without the need to adapt to variations). The method specifies how the fitness of candidate solutions can be evaluated, how the environmental conditions should vary during the course of the evolutionary process, which algorithm can be used, and how the best solution can be identified. The obtained results s ...[more]
Similar Datasets
Project description:In this paper we compare systematically the most promising neuroevolutionary methods and two new original methods on the double-pole balancing problem with respect to: the ability to discover solutions that are robust to variations of the environment, the speed with which such solutions are found, and the ability to scale-up to more complex versions of the problem. The results indicate that the two original methods introduced in this paper and the Exponential Natural Evolutionary Strategy method largely outperform the other methods with respect to all considered criteria. The results collected in different experimental conditions also reveal the importance of regulating the selective pressure and the importance of exposing evolving agents to variable environmental conditions. The data collected and the results of the comparisons are used to identify the most effective methods and the most promising research directions.
| S-EPMC6051599 | biostudies-literature
Project description:We show analytically that training a neural network by conditioned stochastic mutation or neuroevolution of its weights is equivalent, in the limit of small mutations, to gradient descent on the loss function in the presence of Gaussian white noise. Averaged over independent realizations of the learning process, neuroevolution is equivalent to gradient descent on the loss function. We use numerical simulation to show that this correspondence can be observed for finite mutations, for shallow and deep neural networks. Our results provide a connection between two families of neural-network training methods that are usually considered to be fundamentally different.
| S-EPMC8563972 | biostudies-literature
Project description:Robust optimization leads to intensity-modulated proton therapy (IMPT) plans that are less sensitive to uncertainties and superior in terms of organs-at-risk (OARs) sparing, target dose coverage, and homogeneity compared to planning target volume (PTV)-based optimized plans. Robust optimization incorporates setup and range uncertainties, which implicitly adds margins to both targets and OARs and is also able to compensate for perturbations in dose distributions within targets and OARs caused by uncertainties. In contrast, the traditional PTV-based optimization considers only setup uncertainties and adds a margin only to targets but no margins to the OARs. It also ignores range uncertainty. The purpose of this work is to determine if robustly optimized plans are superior to PTV-based plans simply because the latter do not assign margins to OARs during optimization.The authors retrospectively selected from their institutional database five patients with head and neck (H&N) cancer and one with prostate cancer for this analysis. Using their original images and prescriptions, the authors created new IMPT plans using three methods: PTV-based optimization, optimization based on the PTV and planning risk volumes (PRVs) (i.e., "PTV+PRV-based optimization"), and robust optimization using the "worst-case" dose distribution. The PRVs were generated by uniformly expanding OARs by 3 mm for the H&N cases and 5 mm for the prostate case. The dose-volume histograms (DVHs) from the worst-case dose distributions were used to assess and compare plan quality. Families of DVHs for each uncertainty for all structures of interest were plotted along with the nominal DVHs. The width of the "bands" of DVHs was used to quantify the plan sensitivity to uncertainty.Compared with conventional PTV-based and PTV+PRV-based planning, robust optimization led to a smaller bandwidth for the targets in the face of uncertainties {clinical target volume [CTV] bandwidth: 0.59 [robust], 3.53 [PTV+PRV], and 3.53 [PTV] Gy (RBE)}. It also resulted in higher doses to 95% of the CTV {D(95%): 60.8 [robust] vs 59.3 [PTV+PRV] vs 59.6 [PTV] Gy (RBE)}, smaller D(5%) (doses to 5% of the CTV) minus D(95%) {D(5%) - D(95%): 13.2 [robust] vs 17.5 [PTV+PRV] vs 17.2 [PTV] Gy (RBE)}. At the same time, the robust optimization method irradiated OARs less {maximum dose to 1 cm(3) of the brainstem: 48.3 [robust] vs 48.8 [PTV+PRV] vs 51.2 [PTV] Gy (RBE); mean dose to the oral cavity: 22.3 [robust] vs 22.9 [PTV+PRV] vs 26.1 [PTV] Gy (RBE); maximum dose to 1% of the normal brain: 66.0 [robust] vs 68.0 [PTV+PRV] vs 69.3 [PTV] Gy (RBE)}.For H&N cases studied, OAR sparing in PTV+PRV-based optimization was inferior compared to robust optimization but was superior compared to PTV-based optimization; however, target dose robustness and homogeneity were comparable in the PTV+PRV-based and PTV-based optimizations. The same pattern held for the prostate case. The authors' data suggest that the superiority of robust optimization is not due simply to its inclusion of margins for OARs, but that this is due mainly to the ability of robust optimization to compensate for perturbations in dose distributions within target volumes and normal tissues caused by uncertainties.
| S-EPMC3562272 | biostudies-other
Project description:When conducting high-throughput biological experiments, it is often necessary to develop a protocol that is both inexpensive and robust. Standard approaches are either not cost-effective or arrive at an optimized protocol that is sensitive to experimental variations. We show here a novel approach that directly minimizes the cost of the protocol while ensuring the protocol is robust to experimental variation. Our approach uses a risk-averse conditional value-at-risk criterion in a robust parameter design framework. We demonstrate this approach on a polymerase chain reaction protocol and show that our improved protocol is less expensive than the standard protocol and more robust than a protocol optimized without consideration of experimental variation.
| S-EPMC4582800 | biostudies-literature
Project description:The presence of outliers in financial asset returns is a frequently occurring phenomenon which may lead to unreliable mean-variance optimized portfolios. This fact is due to the unbounded influence that outliers can have on the mean returns and covariance estimators that are inputs in the optimization procedure. In this paper we present robust estimators of mean and covariance matrix obtained by minimizing an empirical version of a pseudodistance between the assumed model and the true model underlying the data. We prove and discuss theoretical properties of these estimators, such as affine equivariance, B-robustness, asymptotic normality and asymptotic relative efficiency. These estimators can be easily used in place of the classical estimators, thereby providing robust optimized portfolios. A Monte Carlo simulation study and applications to real data show the advantages of the proposed approach. We study both in-sample and out-of-sample performance of the proposed robust portfolios comparing them with some other portfolios known in literature.
| S-EPMC4607458 | biostudies-literature
Project description:Intensity modulated proton therapy (IMPT) is highly sensitive to range uncertainties and uncertainties caused by setup variation. The conventional inverse treatment planning of IMPT optimized based on the planning target volume (PTV) is not often sufficient to ensure robustness of treatment plans. In this paper, a method that takes the uncertainties into account during plan optimization is used to mitigate the influence of uncertainties in IMPT.The authors use the so-called "worst-case robust optimization" to render IMPT plans robust in the face of uncertainties. For each iteration, nine different dose distributions are computed-one each for ± setup uncertainties along anteroposterior (A-P), lateral (R-L) and superior-inferior (S-I) directions, for ± range uncertainty, and the nominal dose distribution. The worst-case dose distribution is obtained by assigning the lowest dose among the nine doses to each voxel in the clinical target volume (CTV) and the highest dose to each voxel outside the CTV. Conceptually, the use of worst-case dose distribution is similar to the dose distribution achieved based on the use of PTV in traditional planning. The objective function value for a given iteration is computed using this worst-case dose distribution. The objective function used has been extended to further constrain the target dose inhomogeneity.The worst-case robust optimization method is applied to a lung case, a skull base case, and a prostate case. Compared with IMPT plans optimized using conventional methods based on the PTV, our method yields plans that are considerably less sensitive to range and setup uncertainties. An interesting finding of the work presented here is that, in addition to reducing sensitivity to uncertainties, robust optimization also leads to improved optimality of treatment plans compared to the PTV-based optimization. This is reflected in reduction in plan scores and in the lower normal tissue doses for the same coverage of the target volume when subjected to uncertainties.The authors find that the worst-case robust optimization provides robust target coverage without sacrificing, and possibly even improving, the sparing of normal tissues. Our results demonstrate the importance of robust optimization. The authors assert that all IMPT plans should be robustly optimized.
| S-EPMC3281975 | biostudies-other
Project description:PurposeTo develop and validate a robust template for VMAT SBRT of lung lesions, using the multicriterial optimization (MCO) of a commercial treatment planning system.MethodsThe template was established and refined on 10 lung SBRT patients planned for 55 Gy/5 fr. To improve gradient and conformity a ring structure around the planning target volume (PTV) was set in the list of objectives. Ideal fluence optimization was conducted giving priority to organs at risk (OARs) and using the MCO, which further pushes OARs doses. Segmentation was conducted giving priority to PTV coverage. Two different templates were produced with different degrees of modulation, by setting the Fluence Smoothing parameter to Medium (MFS) and High (HFS). Each template was applied on 20 further patients. Automatic and manual plans were compared in terms of dosimetric parameters, delivery time, and complexity. Statistical significance of differences was evaluated using paired two-sided Wilcoxon signed-rank test.ResultsNo statistically significant differences in PTV coverage and maximum dose were observed, while an improvement was observed in gradient and conformity. A general improvement in dose to OARs was seen, which resulted to be significant for chest wall V30 Gy , total lung V20 Gy , and spinal cord D0.1 cc . MFS plans are characterized by a higher modulation and longer delivery time than manual plans. HFS plans have a modulation and a delivery time comparable to manual plans, but still present an advantage in terms of gradient.ConclusionThe automation of the planning process for lung SBRT using robust templates and MCO was demonstrated to be feasible and more efficient.
| S-EPMC7324702 | biostudies-literature
Project description:Promoted model architectures or algorithms are crucial for intelligent manufacturing since developing them takes a lot of trial and error to embed the domain knowledge into the models correctly. Especially in semiconductor manufacturing, the whole processes depend on complicated physical equations and sophisticated fine-tuning. Therefore, we use a neuroevolution-based model to search the optimized architecture automatically. The collector current value at a particular bias of the silicon-germanium (SiGe) heterojunction bipolar transistor, generated by technology computer-aided design (TCAD), is used as the target dataset with six process parameters as the inputs. The processes include oxidation, dry and wet etching, implantation, annealing, diffusion, and chemical-mechanical polishing. Our work can build a suitable model network with a fast turnaround time, and practical physical constraints are fused in it without domain knowledge extraction. Take the case with 3840 data and one output as an instance. The mean square errors of the train set and validation set, as well as the mean absolute percentage error of the test set, are 1.317 × 10-6, 7.215 × 10-7, and 0.216 while using multilayer perceptron (MLP) and they are 3.285 × 10-7, 1.661 × 10-7, and 0.097 while using NE. The consequences show that the work in this vein is promising. According to the trend plot and results, the ability to extract physic is much better than the traditional (MLP) model.
| S-EPMC10413832 | biostudies-literature
Project description:Enzyme catalysts of a retroaldol reaction have been generated by computational design using a motif that combines a lysine in a nonpolar environment with water-mediated stabilization of the carbinolamine hydroxyl and ?-hydroxyl groups. Here, we show that the design process is robust and repeatable, with 33 new active designs constructed on 13 different protein scaffold backbones. The initial activities are not high but are increased through site-directed mutagenesis and laboratory evolution. Mutational data highlight areas for improvement in design. Different designed catalysts give different borohydride-reduced reaction intermediates, suggesting a distribution of properties of the designed enzymes that may be further explored and exploited.
| S-EPMC3403469 | biostudies-literature
Project description:In this work, we argue that the search for Artificial General Intelligence should start from a much lower level than human-level intelligence. The circumstances of intelligent behavior in nature resulted from an organism interacting with its surrounding environment, which could change over time and exert pressure on the organism to allow for learning of new behaviors or environment models. Our hypothesis is that learning occurs through interpreting sensory feedback when an agent acts in an environment. For that to happen, a body and a reactive environment are needed. We evaluate a method to evolve a biologically-inspired artificial neural network that learns from environment reactions named Neuroevolution of Artificial General Intelligence, a framework for low-level artificial general intelligence. This method allows the evolutionary complexification of a randomly-initialized spiking neural network with adaptive synapses, which controls agents instantiated in mutable environments. Such a configuration allows us to benchmark the adaptivity and generality of the controllers. The chosen tasks in the mutable environments are food foraging, emulation of logic gates, and cart-pole balancing. The three tasks are successfully solved with rather small network topologies and therefore it opens up the possibility of experimenting with more complex tasks and scenarios where curriculum learning is beneficial.
| S-EPMC9613950 | biostudies-literature