Kevin Dong - Machine Learning - Self-Assessment

Overview of Things Learned
Technical:

  • NLP annotation tool characteristics
  • Active learning architecture
  • Using Gaussian Processes in conjunction with traditional ML classifiers

Tools:

  • Python
  • Google Colab
  • Github
  • Miro

Soft Skills:

  • Leading a team remotely
  • Conducting remote team meetings, tutorials and brainstorming
  • Remote team motivation and socials

Achievement Highlights

  • Successfully conducted 5 team meetings and retained 100% of team members since the beginning of the July session
  • Developed an active learning architecture with several variants for an NLP annotation tool
  • Implemented professional software development processes within the team, including the structured use of version control, UML diagrams and the agile iterative process (Scrum)

List of Meetings/Training Attended

  • All team meetings, work sessions and socials

Goals for Next Week

  • Complete Iteration 1 of our Scrum workflow and produce a product with basic functionality for benchmarking with Stack Exchange data
  • Investigate the various models/techniques that can be used for tag prediction and decide on which to focus on and assign to members
  • Begin Iteration 2 which should focus on model exploration, benchmarking data analysis and the transition to operate on STEM-Away data

Detailed Statement of Tasks Done

  • To improve engagement during meetings, everyone is given a chance to speak. No matter how small or insignificant an input may seem, everyone’s contributions are welcome. Miro is used as a live whiteboarding tool to provide visuals that complement spoken concepts and ideas. Leads turned on their videos to have a more personal connection with members. Meetings are recorded and meeting summaries are shared on STEM-Away for those who are temporarily absent.
  • Miro’s diagraming tools are leveraged to construct brainstorming boards, architecture diagrams, UML diagrams and a Kanban board to facilitate collaboration and provide structure to our work. Github workflow guidelines are established and team members are briefed on the usage of all collaboration tools and processes. Object-oriented programming principles are adopted to modularize our code to facilitate assignment and evolution.
  • An informal atmosphere is created to improve team dynamics between leads and participants. “Project parties” are advertised as drop-in programming sessions instead of mandatory meetings. Scheduled team socials have not yet been conducted due to the concern that they may conflict with everyone’s busy schedules, but an social topic has been created on STEM-Away for members to engage in non-work activities such as guessing movies from emojis.