On Consensus-Optimality Trade-offs in Collaborative Deep Learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Jiang Z
PROVIDER: S-EPMC8478077 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Value-based decision-making involves trading off the cost associated with an action against its expected reward. Research has shown that both physical and mental effort constitute such subjective costs, biasing choices away from effortful actions, and discounting the value of obtained rewards. Facing conflicts between competing action alternatives is considered aversive, as recruiting cognitive control to overcome conflict is effortful. Moreover, engaging control to proactively suppress irrelevant information that could conflict with task-relevant information would presumably also be cognitively costly. Yet, it remains unclear whether the cognitive control demands involved in preventing and resolving conflict also constitute costs in value-based decisions. The present study investigated this question by embedding irrelevant distractors (flanker arrows) within a reversal-learning task, with intermixed free and instructed trials. Results showed that participants learned to adapt their free choices to maximize rewards, but were nevertheless biased to follow the suggestions of irrelevant distractors. Thus, the perceived cost of investing cognitive control to suppress an external suggestion could sometimes trump internal value representations. By adapting computational models of reinforcement learning, we assessed the influence of conflict at both the decision and learning stages. Modelling the decision showed that free choices were more biased when participants were less sure about which action was more rewarding. This supports the hypothesis that the costs linked to conflict management were traded off against expected rewards. During the learning phase, we found that learning rates were reduced in instructed, relative to free, choices. Learning rates were further reduced by conflict between an instruction and subjective action values, whereas learning was not robustly influenced by conflict between one's actions and external distractors. Our results show that the subjective cognitive control costs linked to conflict factor into value-based decision-making, and highlight that different types of conflict may have different effects on learning about action outcomes.
| S-EPMC6750595 | biostudies-literature
Project description:Proteins are marginally stable, and an understanding of the sequence determinants for improved protein solubility is highly desired. For enzymes, it is well known that many mutations that increase protein solubility decrease catalytic activity. These competing effects frustrate efforts to design and engineer stable, active enzymes without laborious high-throughput activity screens. To address the trade-off between enzyme solubility and activity, we performed deep mutational scanning using two different screens/selections that purport to gauge protein solubility for two full-length enzymes. We assayed a TEM-1 beta-lactamase variant and levoglucosan kinase (LGK) using yeast surface display (YSD) screening and a twin-arginine translocation pathway selection. We then compared these scans with published experimental fitness landscapes. Results from the YSD screen could explain 37% of the variance in the fitness landscapes for one enzyme. Five percent to 10% of all single missense mutations improve solubility, matching theoretical predictions of global protein stability. For a given solubility-enhancing mutation, the probability that it would retain wild-type fitness was correlated with evolutionary conservation and distance to active site, and anticorrelated with contact number. Hybrid classification models were developed that could predict solubility-enhancing mutations that maintain wild-type fitness with an accuracy of 90%. The downside of using such classification models is the removal of rare mutations that improve both fitness and solubility. To reveal the biophysical basis of enhanced protein solubility and function, we determined the crystallographic structure of one such LGK mutant. Beyond fundamental insights into trade-offs between stability and activity, these results have potential biotechnological applications.
| S-EPMC5338495 | biostudies-literature
Project description:Learning is changed behavior following experience, and ubiquitous in animals including plant-inhabiting predatory mites (Phytoseiidae). Learning has many benefits but also incurs costs, which are only poorly understood. Here, we addressed learning, especially its costs, in the generalist predatory mite Amblyseius swirskii, a biocontrol agent of several herbivores, which can also survive on pollen. The goals of our research were (1) to scrutinize if A. swirskii is able to learn during early life in foraging contexts and, if so, (2) to determine the costs of early learning. In the experiments, we used one difficult-to-grasp prey, i.e., thrips, and one easy-to-grasp prey, i.e., spider mites. Our experiments show that A. swirskii is able to learn during early life. Adult predators attacked prey experienced early in life (i.e., matching prey) more quickly than they attacked unknown (i.e., non-matching) prey. Furthermore, we observed both fitness benefits and operating (physiological) costs of early learning. Predators receiving the matching prey produced the most eggs, whereas predators receiving the non-matching prey produced the least. Thrips-experienced predators needed the longest for juvenile development. Our findings may be used to enhance A. swirskii's efficacy in biological control, by priming young predators on a specific prey early in life.
| S-EPMC4804281 | biostudies-other
Project description:Upon activation, the epidermal growth factor receptor (EGFR) phosphorylates tyrosine residues in its cytoplasmic tail, which triggers the binding of Src homology 2 (SH2) and phosphotyrosine-binding (PTB) domains and initiates downstream signaling. The sequences flanking the tyrosine residues (referred to as "phosphosites") must be compatible with phosphorylation by the EGFR kinase domain and the recruitment of adapter proteins, while minimizing phosphorylation that would reduce the fidelity of signal transmission. To understand how phosphosite sequences encode these functions within a small set of residues, we carried out high-throughput mutational analysis of three phosphosite sequences in the EGFR tail. We used bacterial surface display of peptides coupled with deep sequencing to monitor phosphorylation efficiency and the binding of the SH2 and PTB domains of the adapter proteins Grb2 and Shc1, respectively. We found that the sequences of phosphosites in the EGFR tail are restricted to a subset of the range of sequences that can be phosphorylated efficiently by EGFR. Although efficient phosphorylation by EGFR can occur with either acidic or large hydrophobic residues at the -1 position with respect to the tyrosine, hydrophobic residues are generally excluded from this position in tail sequences. The mutational data suggest that this restriction results in weaker binding to adapter proteins but also disfavors phosphorylation by the cytoplasmic tyrosine kinases c-Src and c-Abl. Our results show how EGFR-family phosphosites achieve a trade-off between minimizing off-pathway phosphorylation and maintaining the ability to recruit the diverse complement of effectors required for downstream pathway activation.
| S-EPMC6077704 | biostudies-literature
Project description:The application of artificial intelligence enhances the ability of sensor and networking technologies to realize smart systems that sense, monitor and automatically control our everyday environments. Intelligent systems and applications often automate decisions based on the outcome of certain machine learning models. They collaborate at an ever increasing scale, ranging from smart homes and smart factories to smart cities. The best performing machine learning model, its architecture and parameters for a given task are ideally automatically determined through a hyperparameter tuning process. At the same time, edge computing is an emerging distributed computing paradigm that aims to bring computation and data storage closer to the location where they are needed to save network bandwidth or reduce the latency of requests. The challenge we address in this work is that hyperparameter tuning does not take into consideration resource trade-offs when selecting the best model for deployment in smart environments. The most accurate model might be prohibitively expensive to computationally evaluate on a resource constrained node at the edge of the network. We propose a multi-objective optimization solution to find acceptable trade-offs between model accuracy and resource consumption to enable the deployment of machine learning models in resource constrained smart environments. We demonstrate the feasibility of our approach by means of an anomaly detection use case. Additionally, we evaluate the extent that transfer learning techniques can be applied to reduce the amount of training required by reusing previous models, parameters and trade-off points from similar settings.
| S-EPMC7070423 | biostudies-literature
Project description:Biological robustness is a principle that may shed light on system-level characteristics of biological systems. One intriguing aspect of the concept of biological robustness is the possible existence of intrinsic trade-offs among robustness, fragility, performance, and so on. At the same time, whether such trade-offs hold regardless of the situation or hold only under specific conditions warrants careful investigation. In this paper, we reassess this concept and argue that biological robustness may hold only when a system is sufficiently optimized and that it may not be conserved when there is room for optimization in its design. Several testable predictions and implications for cell culture experiments are presented.
| S-EPMC2913402 | biostudies-literature
Project description:Feeding a growing and increasingly affluent world will require expanded agricultural production, which may require converting grasslands and forests into cropland. Such conversions can reduce carbon storage, habitat provision, and other ecosystem services, presenting difficult societal trade-offs. In this paper, we use spatially explicit data on agricultural productivity and carbon storage in a global analysis to find where agricultural extensification should occur to meet growing demand while minimizing carbon emissions from land use change. Selective extensification saves ? 6 billion metric tons of carbon compared with a business-as-usual approach, with a value of approximately $1 trillion (2012 US dollars) using recent estimates of the social cost of carbon. This type of spatially explicit geospatial analysis can be expanded to include other ecosystem services and other industries to analyze how to minimize conflicts between economic development and environmental sustainability.
| S-EPMC4151777 | biostudies-literature
Project description:Many fundamental choices in life are intertemporal: they involve trade-offs between sooner and later outcomes. In recent years there has been a surge of interest into how people make intertemporal decisions, given that such decisions are ubiquitous in everyday life and central in domains from substance use to climate change action. While it is clear that people make decisions according to rules, intuitions and habits, they also commonly deliberate over their options, thinking through potential outcomes and reflecting on their own preferences. In this Perspective, we bring to bear recent research into the higher-order capacities that underpin deliberation-particularly those that enable people to think about the future (prospection) and their own thinking (metacognition)-to shed light on intertemporal decision-making. We show how a greater appreciation for these mechanisms of deliberation promises to advance our understanding of intertemporal decision-making and unify a wide range of otherwise disparate choice phenomena.
| S-EPMC7147875 | biostudies-literature
Project description:Many have argued that we may be able to extend life and improve human health through hormesis, the beneficial effects of low-level toxins and other stressors. But, studies of hormesis in model systems have not yet established whether stress-induced benefits are cost free, artifacts of inbreeding, or come with deleterious side effects. Here, we provide evidence that hormesis results in trade-offs with immunity. We find that a single topical dose of dead spores of the entomopathogenic fungus, Metarhizium robertsii, increases the longevity of the fruit fly, Drosophila melanogaster, without significant decreases in fecundity. We find that hormetic benefits of pathogen challenge are greater in lines that lack key components of antifungal immunity (Dif and Turandot M). And, in outbred fly lines, we find that topical pathogen challenge enhances both survival and fecundity, but reduces ability to fight off live infections. The results provide evidence that hormesis is manifested by stress-induced trade-offs with immunity, not cost-free benefits or artifacts of inbreeding. Our findings illuminate mechanisms underlying pathogen-induced life-history trade-offs, and indicate that reduced immune function may be an ironic side effect of the "elixirs of life."
| S-EPMC4282086 | biostudies-other
Project description:Rubisco is the primary carboxylase of the Calvin cycle, the most abundant enzyme in the biosphere, and one of the best-characterized enzymes. On the basis of correlations between Rubisco kinetic parameters, it is widely posited that constraints embedded in the catalytic mechanism enforce trade-offs between CO2 specificity, SC/O, and maximum carboxylation rate, kcat,C. However, the reasoning that established this view was based on data from ≈20 organisms. Here, we re-examine models of trade-offs in Rubisco catalysis using a data set from ≈300 organisms. Correlations between kinetic parameters are substantially attenuated in this larger data set, with the inverse relationship between kcat,C and SC/O being a key example. Nonetheless, measured kinetic parameters display extremely limited variation, consistent with a view of Rubisco as a highly constrained enzyme. More than 95% of kcat,C values are between 1 and 10 s-1, and no measured kcat,C exceeds 15 s-1. Similarly, SC/O varies by only 30% among Form I Rubiscos and <10% among C3 plant enzymes. Limited variation in SC/O forces a strong positive correlation between the catalytic efficiencies (kcat/KM) for carboxylation and oxygenation, consistent with a model of Rubisco catalysis in which increasing the rate of addition of CO2 to the enzyme-substrate complex requires an equal increase in the O2 addition rate. Altogether, these data suggest that Rubisco evolution is tightly constrained by the physicochemical limits of CO2/O2 discrimination.
| S-EPMC6686151 | biostudies-literature