Boost Meeting

pKineto §

• Less than 2 minutes LT / 15 minute presentation
  • Content: Boost Conference Manuscript

Boost Conference

• Insights from kineto
  • Surprisingly, they don’t take notes in university
    • I think they should, actually
  • I thought it would be nice if messages could be conveyed from people who speak earlier to those who speak later in a pseudo-synchronous manner
    • Like using sub-audio or something
    • Or maybe using comments like on NicoNico?
  • It might be good to explore the range of lag that humans can tolerate in communication, it could even become a patent or something @tnatsu
  • “Attending Hermione” (from Harry Potter)
  • We might be able to get cooperation from N High @tnatsu

---

• Trust model

• Data-driven: Doing reinforcement learning in a supervised learning-like way

• Making it possible for even copy-paste programmers to do it

• Live coding

  • You can see the performer’s dedication and thought process
  • Creating something that combines live performance and interactivity

• Sharding for blockchain splitting

  • Making it possible to test it
  • Simulating it on the user/TX layer (layer 1.5)

• It’s sad when bugs are inserted in hardware design

  • So we want to test it
  • Existing methods are difficult
  • Fuzzing (trying inputs to see if bugs occur)
    • Coverage-based fuzzing
      • Mutating prior knowledge and adding it to the input set if it is new coverage

• Space debris

  • Reflection of light changes, can estimate movement with light curve
  • Actually launching satellites to test/validate estimation systems
  • It’s difficult to get information, launching is now cheaper apparently

• Robots

  • Using algorithms to design optimized shapes
  • Trying out the generated abstract designs and regenerating them with data
    • Examples of parameters adjusted by algorithms
      • Number of legs, gaps, etc.
  • New shapes may be born with algorithms, optimized for certain environments

• Untrodden Jr.

  • It’s a complex feeling when people who are definitely stronger than you stay as Jr.
    • Scary, haha

• OB/OG lectures

• Maze person

  • Educational significance of mazes

    • Logical thinking, game creation (creativity), etc.

  • Selecting and discarding

    • Untrodden-related people are overwhelmingly experts, so everything seems correct
    • That’s not true, if you accept everything without thinking, your axis will be off

  • Of course, if it’s from a specialist, you should accept it, but if not, you need to select and discard

  • Knowledge and technical opinions should be sought from Untrodden-related people, impressions are not important

  • Test impressions to death, have prototypes used every week

    • It’s okay to ignore those who are not the target audience, in the case of mazes, listen to the children more than the parents

• Hack4Play

  • Wandering is inevitable
  • The more you fail, the more memories you have
  • Once you graduate, you can easily participate in meetings, haha
    • No matter what happens, March will come