Human and Smart Machine Co-Learning

Website: http://oase.nutn.edu.tw/IEEESMC2018/description.htm

1. Schedule

Room: 3F Fuko

Date: Sunday, October 7 (13:00-17:00) and Monday, October 8 (09:30-17:30)

2. Theme and Organizers

  • Theme: Robotic Open Go System and BCI-based Game System for Human Interactive Learning on Cybernetics @ IEEE SMC 2018
  • Organizers
    • Shun-Feng Su, National Taiwan University of Science and Technology, Taiwan
    • Chang-Shing Lee, National University of Tainan, Taiwan
    • Naoyuki Kubota, Tokyo Metropolitan University, Japan
    • Marek Reformat, University of Albert, Canada

3. Program

  • To have a panel on “Future on Smart Machine e-Learning.”
  • To have a special event, “Human Interactive Learning on Cybernetics.”
  • To have three activities for “Robotic Open Go System / BCI-based Serious Games System for Human Interactive Learning on Cybernetics” on site of IEEE SMC 2018.
    • Robotic BCI-DDF Go / BCI-based Serious Game + Human / Taiwan vs. Robotic BCI-DDF Go BCI-based Serious Game + Human / Taiwan
    • Robotic BCI-DDF Go BCI-based Serious Game + Human / Japan vs. Robotic BCI-DDF Go BCI-based Serious Game + Human / Taiwan
    • Robotic BCI-DDF Go BCI-based Serious Game + Human / Taiwan vs. Robotic BCI-DDF Go BCI-based Serious Game + Human / Japan

4. Short Description of Actions

  • Learning has become a very popular approach for cybernetics systems. This topic has always been considered a research in the Computational Intelligence area. Nevertheless, when talking about smart machines, it is not just about the methodologies. We need to consider systems and cybernetics. Sometimes, we also need to include human in the loop. Thus, it is definitely a research topic in SMC society. About the series game, an intelligent robot agent based on domain ontology, machine learning mechanism, and Fuzzy Markup Language (FML) for students and robot co-learning is presented. The machine-human co-learning model is established to help various students learn the mathematical concepts based on their learning ability and performance. Meanwhile, the robot acts as a teacher’s assistant to co-learn with children in the class. The FML-based knowledge base and rule base are embedded in the robot so that the teachers can get feedback from the robot on whether students make progress or not. Next, we infer students’ learning performance based on learning content’s difficulty and students’ ability, concentration level, as well as teamwork spirit in the class. Moreover, we combine the optimization techniques such as genetic algorithm (GA) and particle swarm optimization (PSO) with FML, called GFML and PFML, respectively, to learn the constructed knowledge base and rule base. Experimental results show that learning with the robot is helpful for disadvantaged and below-basic children [6].
  • The brainwave technology has been developed for a long time; however, applying it to play Go is the world’s first case in an IEEE conference. The world latest mobile and wireless EEG system is fully utilized in the innovation of the developed BCI-DDF Go system. The wireless system, developed by the research team from Brain Research Center in NCTU, is designed to extract the Go player’s brainwaves when they play and compete with the DDF Go system directly [3].
  • In the special event of IEEE SMC 2018 (http://oase.nutn.edu.tw/IEEESMC2018/description.htm), we will combine the theory of deep learning with the technology of BCI [1, 2] to demonstrate playing Go via cellphone.
    • Steady-State Visual Evoked Potential (SSVEP) is a brain activity in response to a rapid and repetitive visual stimulus flashing higher than 3Hz. SSVEP-based brain-computer interface (BCI) systems have attracted a lot of attention due to the high information transfer rate, ease-of-use, and less training time. In general, a SSVEP-based BCI system is consisted of three components: a display, a signal-processing platform, and an Electroencephalogram (EEG) acquisition headset. Display presents single or multiple frequency-tagged visual stimuli to code users’ intention; a signal-processing platform extracts informative features to decode users’ intention; an EEG acquisition headset obtains brain activities from scalp. Although the architecture is simple, moving the entire system that is with bulky display and wired EEG headset from a well-controlled laboratory to a real-world application still poses a lot of challenges. To this end, a portable platform to present accurate visual stimulus and a wireless EEG headset is crucial.
    • This demonstration implements an approximate visual stimulus algorithm in an off-the-shelf smartphone (Samsung Galaxy S8). Five virtual buttons, UP, DOWN, LEFT, RIGHT, and ENTER, flashing at 7, 8, 9, 10, and 11 Hz on the screen simultaneously. When a user gazes at one button for few seconds, it modulates brain activities to the same frequency. For instance, when a user gazes at UP button for few seconds, a 7 Hz peak can be obtained in the frequency domain using FFT. In the other hand, this project uses a 4-channel wireless EEG headset to obtain brain activities. The EEG headset is consisted of an ADC module, amplifiers, and a wireless module with sampling rate of 128 Hz. The dry and spring-loaded electrodes shorten the experimental preparation time with acceptable signal-to-noise ratio. The near real-time brain activities were obtained and sent to a laptop to extract features and decode the user’s intention. For instance, the laptop returns UP to the GO system to move the cursor up using web socket protocol.
    • Possible applications: Since SSVEP has relative higher signal-to-noise ratio and robust data quality compared to other type of EEG signal, it has been applied in many areas. For instance, studies have reported the performance of retrieval of visual working memory can be assessed by computing the amplitude of SSVEP during encoding session. In addition, spelling using SSVEP also showed the highest information transfer rate (over 300 bits/minute) among other EEG signal.
  • In addition to Go system, we will hold Robotic BCI-based Game System for Human Interactive Learning on Cybernetics in IEEE SMC 2018. Meanwhile, we also have an associated special session on Semantic Web Technologies and Ontology for Real-World Applications and a panel on Future Smart Human e-Learning to attract / encourage more researchers and scholars to submit their valuable papers to IEEE SMC 2018, to attend IEEE SMC 2018, to join SMC conference, and then to join the SMC society in the future.
  • We got the 3-year research project (Intelligent IRT Robot and Humans Co-Learning on Education and Learning Applications) from Ministry of Science and Technology (MOST, Taiwan). With the MOST research project fund, we will continue further to enhance the co-learning between humans and robots and will partially support the held activities in IEEE SMC 2018. Now, we already successfully integrated Facebook ELFOpenGo [7] with our developed system. We will have a demonstration at the special event of IEEE SMC 2018.
  • Smart machine is one of the important themes of IEEE SMC Society. It is good to use this competition of Professional Players vs. Machine and also have some cooperation between them to attract more attentions of worldwide scholars to SMC conferences. It definitely will have more papers submissions in this area to IEEE SMC 2018 and more researchers to join SMC society.

Reference

  1. T. Lin, Y. T. Liu, S. L. Wu, Z. Cao, Y. K. Wang, C. S. Huang, J. T. King, S. A. Chen, S. W. Lu, and C. H. Chuang, “EEG-based brain-computer interfaces,” IEEE Systems, Man, and Cybernetics Magazine, vol. 3, no. 1, pp. 16-26. Oct. 2017.
  2. P. Thomas and A. P. Vinod, “Toward EEG-based biometric systems,” IEEE Systems, Man, and Cybernetics Magazine, vol. 3, no. 1, pp. 6-15. Oct. 2017.
  3. S. Lee, M. H. Wang, L. W. Ko, N. Kubota, L. A. Lin, S. Kitaoka, Y. T Wang, and S. F. Su, “Human and smart machine co-learning: brain-computer interaction at the 2017 IEEE International Conference on Systems, Man, and Cybernetics,” IEEE Systems, Man, and Cybernetics Magazine, vol. 4, no. 2, pp. 6-13, Apr. 2018.
  4. Gibney, “Google secretly tested AI bot: updated version of Google DeepMind’s AlphaGo program revealed as mystery online player,” Nature, vol. 541, pp. 142, Jan. 2017.
  5. S. Lee, M. H. Wang, C. S. Wang, O. Teytaud, J. L. Liu, S. W. Lin, and P. H. Hung, “PSO-based fuzzy markup language for student learning performance evaluation and educational application,” IEEE Transactions on Fuzzy Systems, 2018. (DOI: 10.1109/TFUZZ.2018.2810814)
  6. S. Lee, M. H. Wang, T. X. Huang, L. C. Chen, Y. C. Huang, S. C. Yang, C. H. Tseng, P. H. Hung, and N. Kubota, “Ontology-based fuzzy markup language agent for student and robot co-learning,” 2018 World Congress on Computational Intelligence (IEEE WCCI 2018), Rio de Janeiro, Brazil, Jul. 8-13, 2018.
  7. D. Tian and L. Zitnick, “Facebook Open Sources ELF OpenGo,” May 2018, [Online] Available: https://research.fb.com/facebook-open-sources-elf-opengo/.