VIP20
本课程属于：035-AI for CAE Simulation GNN and VVUQ English Technical Courses · 共 20 门相关课程，组合学习更完整。
查看所属VIP全部课程 →
← 返回所属目录 咨询 / 购买

Bayesian Multi-Objective Optimization and Trusted AI Deployment

30 Chapters
View Course Samples Online
01
Course overview
scope, concepts, and technical context of Bayesian multi-objective optimization and trusted AI deployment.
technicalBayesian
02
Multi-objective optimization fundamentals
Pareto dominance, Pareto front, hypervolume, and performance metrics.
Multi-objectiveoptimization
03
Bayesian optimization foundations
Gaussian processes as surrogate models, acquisition functions, and the exploration-exploitation trade-off.
Bayesianoptimization
04
Bayesian multi-objective optimization
expected hypervolume improvement, Pareto front modeling, and scalarization-free approaches.
Bayesianmulti-objective
05
Acquisition functions for multi-objective settings
EHVI, ParEGO, SMS-EGO, and uncertainty-aware selection.
Acquisitionfunctions
06
Batch and asynchronous Bayesian multi-objective optimization
parallel evaluations, delayed observations, and resource constraints.
asynchronousBayesian
07
Constrained Bayesian multi-objective optimization
feasibility modeling, chance constraints, and constrained acquisition functions.
ConstrainedBayesian
08
High-dimensional and structured Bayesian multi-objective optimization
dimensionality reduction, additive models, and trust regions.
High-dimensionalstructured
09
Multi-fidelity Bayesian multi-objective optimization
fidelity-aware surrogates, cost-aware acquisition, and budget allocation.
Multi-fidelityBayesian
10
Preference elicitation and incorporation in Bayesian multi-objective optimization
utility functions, preference learning, and interactive methods.
Preferenceelicitation
11
Robust and stochastic Bayesian multi-objective optimization
noise handling, input uncertainty, and robust Pareto optimality.
stochasticBayesian
12
Bayesian multi-objective optimization for hyperparameter tuning
multi-metric model selection, early stopping, and resource-aware search.
Bayesianmulti-objective
13
Bayesian multi-objective optimization in engineering design
simulation-based optimization, trade-off analysis, and design space exploration.
Bayesianmulti-objective
14
Bayesian multi-objective optimization for materials and drug discovery
experimental design, batch synthesis, and active learning.
Bayesianmulti-objective
15
Bayesian multi-objective optimization in control and robotics
policy search, safety constraints, and real-time adaptation.
Bayesianmulti-objective
16
Trusted AI deployment overview
trustworthiness dimensions, risk management, and lifecycle governance.
deploymenttrustworthiness
17
Uncertainty quantification for trusted deployment
calibration, Bayesian neural networks, and predictive uncertainty in decision-making.
Uncertaintyquantification
18
Robustness and generalization in deployment
distribution shift, adversarial robustness, and stress testing.
Robustnessgeneralization
19
Fairness and bias mitigation in deployed AI systems
fairness metrics, bias detection, and mitigation strategies.
Fairnessmitigation
20
Explainability and interpretability for trusted AI
post-hoc explanations, inherently interpretable models, and explanation evaluation.
Explainabilityinterpretability
21
Privacy-preserving AI deployment
differential privacy, federated learning, and secure multi-party computation.
Privacy-preservingdeployment
22
Security of AI systems
adversarial attacks, model stealing, data poisoning, and defenses.
Securityadversarial
23
Safety assurance and verification
formal methods, runtime monitoring, and safety cases for AI systems.
assuranceverification
24
Model monitoring and maintenance in production
drift detection, performance degradation, and continual learning.
monitoringmaintenance
25
Deployment architectures for trusted AI
edge, cloud, and hybrid; model serving, versioning, and rollback.
Deploymentarchitectures
26
Regulatory and ethical frameworks for AI deployment
GDPR, AI Act, accountability, and transparency requirements.
Regulatoryframeworks
27
Integrating Bayesian multi-objective optimization with trusted AI deployment
optimizing for trust metrics, trade-offs, and constraints.
IntegratingBayesian
28
Case study
trusted AI deployment in healthcare—Bayesian multi-objective optimization for treatment policies with safety and fairness constraints.
deploymenthealthcare
29
Case study
trusted AI deployment in autonomous systems—multi-objective optimization for perception and control with robustness guarantees.
deploymentautonomous
30
Capstone project
end-to-end Bayesian multi-objective optimization and trusted deployment pipeline for a real-world application.
Capstoneend-to-end