At ZJUI, there is a course that is highly favored by students in Computer Engineering, Electrical Engineering, and Mechanical Engineering for its innovative and unique course content and superior course design. At the same time, students in Bioinformatics and Business also choose the course across institutes, making it one of the most popular courses among ZJUI students. ECE 448 initially originated from a collaborative teaching project with the University of Illinois Urbana Champaign (UIUC), gradually enriching by incorporating more cutting-edge technological elements, interdisciplinary knowledge, and practical application examples as the course content. Recently, this course was successfully approved as a university-level AI For Education series empirical teaching research project.

ECE 448 is a course taught by ZJUI Professor Wang Hongwei, which introduces the basic theory, key technologies, and applications of artificial intelligence in multiple professional fields. The content covers search algorithms, Bayesian inference, machine learning and reinforcement learning, natural language processing, etc. This course provides students with a unique classroom experience through its distinct interdisciplinary features, strong interest, and innovative teaching methods.

 

 

Interdisciplinary Learning: Rooted in Diversity

 

The course creatively leverages an interdisciplinary project-based learning model empowered by Generative Artificial Intelligence (GAI). It vigorously motivates students to skillfully transfer and apply AI knowledge to other multiple disciplines like Biology, Economics, and Psychology. Through this innovative learning process, students gradually hone their exceptional abilities to utilize AI technologies to analyze and solve various complex problems extensively and accurately. Moreover, this unique learning approach significantly stimulates students' creative thinking and critical analysis abilities. It kindles a continuous stream of innovative sparks through the collision and integration of knowledge.

 

 

▲ Example 1：AI application in automatic driving

 

 

▲ Example 2：AI application in the medical field

 

 

▲ Example 3：Earthquake prediction with Bayesian networks

 

 

▲ Example 4：Stock market trend prediction using deep learning

 

 

AI Learning Assistant: Powering Support

 

The course innovatively incorporates GAI into teaching, enabling AI to serve as 7x24-hour intelligent learning assistants for students. Inspired by the course, students have gradually started to use generative AI tools, represented by ChatGPT, to assist in reviewing and consolidating the concepts learned in class, thereby deepening their understanding and memory of the knowledge acquired. Additionally, the AI systems can realistically simulate various learning scenarios, assisting students in overcoming challenging and complex concepts, establishing clearer knowledge networks, and thoughtfully offering emotional support and care, making the learning journey be not only full of knowledge acquisition but also filled with the joy and efficiency of learning.

 

▲ Using ChatGPT to reinforce the concepts of breadth-first search algorithm

 

▲ Using ChatGPT to reinforce the concepts of reinforcement learning

 

 

Fun Learning: Search Algos for Maze Solving

 

The assignments for ECE 448 are AI projects which are thoughtfully designed and closely connected to real-world scenarios, such as building image recognition models or designing maze pathfinding algorithms, allowing students to apply theoretical knowledge in practical contexts in an engaging and entertaining way. During this process, students are required to leverage practical programming projects to apply theoretical knowledge to real-world problem-solving. Taking the virtual "Pacman" character as an example, for which students need to design a maze pathfinding program. In the maze scenario, Pacman starts at a certain position, eats many food particles in sequence, and reaches the target position. Along the way, students should choose a proper optimal search strategy among various search strategies to achieve the shortest path. These strategies include Depth-First Search (DFS), Breadth-First Search (BFS), and Greedy Best-First Search, etc. Through such projects, students not only master programming skills but also gain a profound understanding of the entire process and the elegance of algorithm design and problem-solving.

 

▲ Maze Path Planning Program Design

 

 

 

Interview Time:

Let’s listen to what the professor said.

 

01 How is the course designed?

 

The design of this course adheres to the core philosophy of balancing theory and practice, aiming to provide students with a solid foundation in theoretical knowledge while enabling them to effectively tackle complex real-world problems using artificial intelligence techniques through hands-on practice. The course content covers a wide range of fundamental concepts, frameworks, and core technologies in AI, deeply exploring the innovative applications of these technologies in various fields, such as search algorithms, classification techniques, logical reasoning mechanisms, machine learning paradigms, reinforcement learning strategies, computer vision technologies, and natural language processing methods.

 

The course aims to help students develop the ability to proficiently use programming to solve real-world problems after completing their studies successfully, while fostering critical thinking and deep analytical skills. Through hands-on programming exercises and professional assessments, students will engage in practical scenarios to effectively hone and demonstrate these skills. Additionally, the course is taught entirely in English, encouraging students to communicate in English both inside and outside the classroom, thereby cultivating their global perspective and cross-cultural adaptability. Ultimately, this course guides students in developing interdisciplinary innovative thinking, enabling them to apply it to innovative practices in engineering technology.

 

02 What do you think are the main challenges of the course?

 

Firstly, the complexity of artificial intelligence can’t be underestimated. Topics like search algorithms, machine learning models are highly abstract and conceptually profound. When integrated with mathematical derivation and programming, the high requirements for mathematical and programming skills make it difficult for many students to get started.

 

Secondly, some students have a solid foundation and quickly grasp the knowledge, while others may lack a basic understanding or struggle with foundational concepts. Traditional 'one-size-fits-all' teaching methods are unable to meet the diverse needs of students at varying levels. So, it is necessary to design personalized instructional content to accommodate different learning paces.

 

Furthermore, balancing the theoretical and practical aspects of teaching is quite challenging. AI requires both a deep understanding of theory and extensive practical application. How to balance both within limited time is a significant challenge for course design and implementation.

 

03 What teaching methods or strategies do you primarily use to convey knowledge?

 

1.Project-Based Learning: Through hands-on programming projects, students are guided to apply theoretical knowledge to real-world problems.

 

2.Case-Based Teaching: By analyzing classic AI cases (such as AlphaGo, ChatGPT, autonomous driving, etc.), students gain insight into AI applications and their societal impacts.

 

3.Hands-On Experiments and Programming Practice: Designing programming tasks of various difficulty levels, starting from data preprocessing to constructing neural network models, guides students to grasp the core elements of AI skills gradually and cultivates their abilities of programming and implementing algorithms.

 

4.Guided Learning: By encouraging students to ask questions actively and explore solutions independently,  this approach fosters their ability for self-directed learning while progressively refining their model design.

 

5.Group Collaboration: teamwork and competition can greatly stimulate students’ learning enthusiasm.

 

04 Message to Students

 

Welcome to the world of artificial intelligence! This is a field filled with endless possibilities and creativity that are bound to redefine our understanding of life, work, and the future. Whether you are new to AI or have some prior knowledge, today marks an important starting point for you to explore this vast domain.

 

In the journey of learning this course, you’ll realize that AI is not just merely a stack of coding and algorithms, but it is a bridge between human intelligence and technology. It requires rigorous logical thinking as well as imaginative leaps. We study AI not only to gain a skill but to explore how to use technology to solve real-world problems, improve lives, and create value.

 

As your instructor, I encourage everyone to:

1.Maintain Curiosity

Keep asking "why" and "what if." These questions are the driving force behind technological advancements.

2.Be Bold in Trial and Error

The process of coding and optimizing algorithms is filled with challenges, but each step of debugging and refining brings you closer to success.

3.Collaboration and Sharing

AI is an interdisciplinary field. Collaborating and sharing knowledge will greatly benefit your learning journey.

4.Consider the Responsibility of Technology

AI is a double-edged sword. How we can make technology serve humanity is a crucial question for everyone.

 

 

Let’ s listen to what the teaching assistant said.

 

Tao Shuting – a 2021 doctoral student in Computer Science and Technology

 

As a teaching assistant for this course, I am keenly aware of the challenges and significance inherent in the course design and implementation. Firstly, the course content is extensive and in-depth, covering various fields of artificial intelligence such as search algorithms, machine learning, computer vision, and natural language processing. For students, the theoretical aspects may initially seem challenging, especially when mathematical derivations and programming practices are integrated. Various abstract concepts take a lot of time for beginners to understand and grasp. As a teaching assistant, my role is to support students in overcoming these challenges. By addressing their queries and uncertainties, I strive to enable them to gain a more profound understanding of theoretical knowledge. Moreover, I aim to make them fully aware of its crucial significance and value in practical scenarios.

 

 

Let’s listen to what the students said.

 

Tang Dianxing – 26’ Undergraduate student in Computer Engineering

 

"Introduction to Artificial Intelligence", taught by Professor Wang Hongwei, is an incredibly beneficial and engaging course. With his deep expertise and engaging teaching style, Professor Wang, simplifies complex AI concepts, making it easier for us to understand the core principles of this cutting-edge field.

 

In the course, Professor Wang not only thoroughly explains the fundamental theories of AI, such as game theory, machine learning, and deep learning, but also through practical case studies, we gain a vivid understanding of AI's extensive applications in the real world—ranging from healthcare to autonomous driving, and smart homes. This teaching method of combining theory and practice greatly enhances our interest in learning AI.

 

Meanwhile, Professor Wang encourages innovative thinking and thoughtfully organizes group discussions and practical projects that not only improve our teamwork skills but also develop our abilities to solve real-world problems. Additionally, he shares the latest research findings and emerging technological trends, significantly broadening our cognitive horizons and knowledge boundaries.

 

"Introduction to Artificial Intelligence" has not only helped us build a solid knowledge foundation, but more importantly, ignited our passion for exploring the field of AI. Whether you are an AI novice or an AI expert, this course can be incredibly rewarding for everyone.

 

 

Wu Qiyang – 26’ Undergraduate Student in Electrical Engineering

 

ECE 448 Introduction to Artificial Intelligence systematically guides us through core AI concepts, supported by practical applications such as game theory, Markov decision processes, and Bayesian inference, laying a solid foundation for understanding AI principles and implementation methods.

 

One particularly noteworthy aspect is the course integrating cutting-edge Generative AI (GAI) technologies which keeps pace with the latest trends. By studying natural language processing (NLP) and deep learning, we gain insights into the basic theories and applications of modern generative models like GPT and deep neural networks. Additionally, the course includes discussions on reinforcement learning and its social impact, stimulating us to explore the practical applications and potential challenges of AI in society.

 

Moreover, the course is well-organized, balancing theory and practice. Through algorithmic operations and project exercises, I have not only gained AI techniques but also developed the ability to solve real-world problems. For those students who are aspiring to pursue AI development, this course is undoubtedly a valuable learning experience.