1. Efficient Multi-Period Trading and Learning Techniques for Portfolio Optimization Open Access Author: Diwan, Ishaan Title: Efficient Multi-Period Trading and Learning Techniques for Portfolio Optimization Area of Honors: Computer Science Keywords: machine learningportfolio optimizationstockstradingboosted decision treesdecision treesMarkowitz mean-variance portfolio theory File: Download Diwan_Thesis_Final.pdf Thesis Supervisors: Mehrdad Mahdavi, Thesis SupervisorJohn Joseph Hannan, Thesis Honors Advisor
2. Online Learning: A Survey Open Access Author: Wang, Weiqin Title: Online Learning: A Survey Area of Honors: Computer Science Keywords: Online LearningMachine Learning File: Download Thesis_Final_Wang.pdf Thesis Supervisors: Mehrdad Mahdavi, Thesis SupervisorTing He, Thesis Honors Advisor
3. PairBoost: Gradient Boosted Classification from Pairwise Data Open Access Author: Ashtekar, Neil Title: PairBoost: Gradient Boosted Classification from Pairwise Data Area of Honors: Computer Science Keywords: Machine LearningEnsemble LearningGradient BoostingData ScienceArtificial Intelligence File: Download Ashtekar_Neil_PairBoost.pdf Thesis Supervisors: Mehrdad Mahdavi, Thesis SupervisorRebecca Jane Passonneau, Thesis Honors Advisor
4. Improving Meta-Learning Performance Through Task Sampling Techniques Open Access Author: Smith, Jacob C Title: Improving Meta-Learning Performance Through Task Sampling Techniques Area of Honors: Computer Science Keywords: Machine LearningMeta-Learning File: Download JacobSmithHonorsThesisFinal.pdf Thesis Supervisors: Mehrdad Mahdavi, Thesis SupervisorDanfeng Zhang, Thesis Honors Advisor
5. Adaptive Partial Training for Model-Heterogeneous Federated Learning Open Access Author: Swope, Jason Title: Adaptive Partial Training for Model-Heterogeneous Federated Learning Area of Honors: Computer Science Keywords: Machine LearningDistributed Machine LearningFederated LearningModel-Heterogeneous Federated LearningPartial TrainingModel HetegeneityData Heterogeneity File: Download Swope_Jason_Adaptive_Partial_Training_for_Model-Heterogeneous_Federated_Learning.pdf Thesis Supervisors: Mehrdad Mahdavi, Thesis SupervisorMohamed Khaled Almekkawy, Thesis Honors Advisor