Machine Learning, Reinforcement Learning, C++, Python, SQL, excel
Resume:
****************@*****.***
Gò Vấp, Hồ Chí Minh
EDUCATION
Hanoi University of Science and
Technology (HUST)
(2021 - 2025)
Bachelor of Electronic and
Telecommunication
CPA: 2.59/4.0
SKILLS
Language
• English - TOEIC 845/990
• German - B1
Machine Learning
• Solid knowledge of Q learning,
MDP formulation.
• Reward Design, Policy choosing.
• Result simulation and evaluation.
Deep Learning
• Supervised and Semi-supervised
models
• Graph neural networks (GCN, GAT)
Programming and tool
• C/C++, Python, SQL
• Microsoft Excel, Word, Powerpoint
CERTIFICATIONS
2024
Business Administrantion
OBJECTIVE
Recent Hanoi University of Science and Technology graduate with a strong passion for AI—especially Machine Learning. Backed by two publications, I’m confident in turning research ideas into well-designed experiments and clear results. I’m eager to learn fast, adapt to new environments, and contribute to teams building practical ML systems.
PROJECTS
Semi-Supervised Broadcast Scheduling in IoT using Graph Convolutional Networks
Paper 1/2025 - 6/2025
• Reframed multi-hop WSN broadcast scheduling as a node time-slot classification task and trained a semi-supervised GCN using features (hop distance, children, descendants ) with a teacher–student loss, achieving up to 99% schedule accuracy.
• My contributions: Co-defined the problem and surveyed solutions; designed the training approach; implemented data pipeline & training scripts and ran experiments; presented and documented key challenges and results. Broadcast scheduling in IoT using Graph Neural Networks with Q - learning generated transmission times
Graduation Thesis 5/2025 - 7/2025
• Graduation Thesis : Built on the prior paper’s WSN/IoT broadcast-scheduling work, reframing the same problem but using Machine Learning to handle labels
—specifically, Q-learning generates transmission-time labels which a Graph Neural Network learns to predict as per-node time slots.
• My contributions: Defined the MDP/env and implemented Q-learning to produce labels; curated graph datasets across topologies and cleaned/aligned labels with slot constraints; ran experiments and ablations vs greedy/SPT baselines; documented challenges and presented the thesis defense. A Q-Learning-Based Broadcast Scheduling Approach for Multi-hop Wireless Sensor Networks
Paper 5/2025 - 10/2025
• Formulated multi-hop WSN broadcast scheduling as a time-slot MDP and proposed a Q-learning policy that selects a frontier transmitter each slot to minimize completion time. Used ε-greedy and latency-aware reward shaping; enforced network; achieved consistent delay reduction vs. greedy baselines across diverse topologies.
Nguyễn Văn Hiếu
AI Engineer Intern
INTERESTS
Badminton - Cycling - Cat - Cafe
• My contributions (end-to-end): Defined the problem & MDP, built the simulator/dataset, designed reward and exploration, implemented Q-learning with logging/visualization, ran experiments & ablations, tuned hyperparameters, analyzed results, and wrote/edited the entire paper to submission. ACTIVITIES
Technology lab on Signal Processing for 5G system
Member 2024 - Now
Research member of the research group on the application of artificial intelligence to wireless communication problems.
© topcv.vn
Contact this candidate