Challenges in Quantum Education

Quantum computing and cryptography are traditionally considered difficult subjects for the students to learn. First, I believe one possible reason for this is that material is presented in a fragmented manner using classroom presentations where the students are treated as passive recipients. Such a set up provides students with limited opportunities to learn the discipline in a holistic manner. Second, quantum computing and cryptography lie at the intersection of several disciplines that includes Physics, Computer Science, Electrical Engineering, Mathematics, as well as Computer Architecture and Hardware. This presents a high bar for the students in terms of pre-requisites. Third, most universities cannot host quantum computers locally and cannot afford quantum cryptography equipment. Furthermore, it requires skilled and trained lab technicians to maintain and run such equipment and does not scale to classroom level learning. And last, internship opportunities in this field are hard to come by. As a result, there is an acute workforce shortage in the field of Quantum Information Sciences (QIS) and the United States is falling behind countries like China, Canada and the European Union.

My lab is confronting these challenges through multiple innovative projects.

Machine Learning Augmented Serious Games

Game-based learning environments, in the area of cybersecurity, have proven to meet student learning objectives and have been popular for decades. They have also become popular in other fields such as medicine, physics, chemical engineering, and even social sciences such as sensitivity and implicit bias training. Such game-based learning environments also provide an alternative means of scalable hands-on holistic experience that is difficult at most universities in the field of quantum computing and cryptography.

My NSF SaTC funded project, QuaSim (QuaSim Website), is a pedagogical game-based virtual educator that allows students an interactive experience by transforming subject-based lectures in quantum cryptography into project-based virtual interactive simulations. While not designed to replace a traditional classical cryptography course, the simulator eases the transition through interactions, visualizations, and game-based scenarios from “classical thinking” to “quantum thinking”.

The project identified and codified essential quantum cryptography knowledge components using first-order logic, developed a concept hierarchy for quantum cryptography, and gamified problems based on the concepts. We also devised algorithms to embed these problems into the chosen QuaSim game assets to provide a coherent gaming experience to the players. Starting from basic concepts such as programming qubits by orienting photons, bases, superposition, measurement, transmission, and reception of qubits to multi-player BB84 quantum key distribution protocol including attacks by an eavesdropping player were developed using this approach. The multi-player version of QuaSim allows three players (Alice, Bob and Eve) to work on exchanging a secret key using the quantum BB84 protocol, each working on a separate machine while sharing a common virtual play space. Furthermore, QuaSim is a media-rich 3D game that employs audio and video narrations, embedded quizzes, exercises, as well avatars to aid players in different scenarios.

Through this project, we also developed framework to instrument QuaSim to collect data on-the-fly about user interactions with problems and game elements and analyzed the data to dynamically adapt the play scenarios. QuaSim records a variety of data such as player use of narrations, performance in tests and quizzes, chats, number of attempts and time taken to solve a problem as well as fine-grained spatial data such as player coordinates, rotation to determine their orientation as well as the use of first and third person views. The data were analyzed to identify player clusters and group them into categories such as hint-seeking, guess-and-check, proficient, players and adapt QuaSim accordingly. Achieving concept proficiency while maintaining user engagement is one of the core tenets of a serious game such as QuaSim. A novel metric combining concept freshness for player engagement while minimizing the time to completion of all scenarios to achieve concept proficiency was developed and used in QuaSim and used to suggest next best scenarios for players in the game. This mechanism was incorporated into QuaSim in automatic, on-demand hint modes in QuaSim to study its effectiveness across different user groups. QuaSim has been used to train over 250 students in quantum computing and cryptography.

While serious games, such as QuaSim, have shown great success in cybersecurity, their applicability has been to a limited audience since not everyone is used to a gaming environment and may even feel uncomfortable in it. As a result, integrating serious games into a traditional classroom has remained a serious challenge.

Intelligent Platforms for Quantum Computing and Cryptography Education

I am currently working on a new adaptive platform, Galore, that combines the strengths of traditional classroom learning teaching with serious games in a seamless manner. Built on the theory of learning objects and Kolb’s experiential learning, Galore allows an educator to create learning objects for different concepts (test case is quantum computing). These learning objects are then instrumented with metadata that allows Galore to search for these objects based on interactivity level, learning preferences such as text, example, widgets, simulations, quizzes or game snippets and automatically synthesizes a family of lesson plans for students. Students log into the system by providing their background information and learning objectives they would like to achieve. Galore then uses a learning object dependency graph to retrieve appropriate learning objects from the learning objects repository and produces lesson plans using interactive Jupyter notebooks as front end. Game snippets are new learning objects that have small focused 2–3-minute game scenarios built into the flow of the lesson plan and aim to provide hands-on lab like environments. No such integration has ever been done before. Galore monitors student performance and may suggest corrective measures by presenting newly retrieved learning objects to students.