GCI World 2026 September Session8 Opening — Transcript
Full transcript
- 0:03Welcome everyone to GCI world session 8.
- 0:06Today's topic is feature engineering and
- 0:10ensemble learning.
- 0:15To begin with, we have a few
- 0:17announcements to share.
- 0:23First, regarding the final assignment,
- 0:25the deadline is November 28 at
- 0:2811:30 a.m. UTC. We expect you to submit
- 0:33your slides, a Jupyter notebook, and a
- 0:35list of references.
- 0:37You may submit as many times as you
- 0:39like. Only the final submission will be
- 0:41graded. Please note that late
- 0:44submissions will not be accepted. So, we
- 0:47recommend submitting early and
- 0:49resubmitting if you make updates.
- 0:52Last week, we held a in-class tutorial
- 0:55and Q&A session for the final
- 0:56assignment. Please review them. Today,
- 0:59read the tutorial and try your first
- 1:01submission.
- 1:03Next, the competition. The deadline is
- 1:06November
- 1:0820
- 1:10uh
- 1:10at
- 1:111:00 a.m. UTC time. Submit the
- 1:15prediction results from your own
- 1:17predictive model. If you are aiming for
- 1:20honors or outstanding student, also
- 1:22submit your code in zip format. You may
- 1:24submit as many times as you like. Only
- 1:27your last submission will be graded. No
- 1:30late submissions.
- 1:31The in-class tutorial and Q&A session
- 1:33were in session 7. Please review them
- 1:35and try to submit today.
- 1:40One important notice about the
- 1:43competition. Please submit the
- 1:45reproducible code you used. Your code
- 1:47must be executable on Google Colab. If
- 1:50you fail to submit your code, your
- 1:53competition score may be invalidated.
- 1:55The deadline is the same as the
- 1:57competition.
- 2:00Uh that's all the announcements. Now,
- 2:02let's kick off session eight.
- 2:05Let me briefly introduce myself. My
- 2:07name's Dean, today's instructor. I'm a
- 2:09master student at the University of
- 2:11Tokyo, and I joined Matsuo Lab at the
- 2:13end of uh 2024. I've worked on
- 2:16enterprise AI and
- 2:18and I'm also a researcher at EPFL in
- 2:21Switzerland. My research interests are
- 2:23AI safety, learning science, human-AI
- 2:25interaction, and uh I'm
- 2:26interpretability. My hobby is cooking.
- 2:29Nice to meet you all.
- 2:31Here's uh today's plan, uh 90 minutes in
- 2:35total.
- 2:37Here's the course road map. We started
- 2:39with the basics and then moved onto
- 2:42machine learning.
- 2:43Uh in parallel, the competition and the
- 2:45final assignment are running along with
- 2:47the side track and business lectures.
- 2:51Today, we are here, uh session eight,
- 2:53feature engineering. Now that you can
- 2:55build and evaluate the model, the next
- 2:58step is learning how to improve it.
- 3:02Uh
- 3:03so today's lecture, feature engineering
- 3:06and ensemble learning.
- 3:09Uh the goal today is to learn how to
- 3:10improve the model you built. First,
- 3:13feature engineering.
- 3:14Uh features need to be pre-pro-
- 3:17pre-processed before the model uses
- 3:20them. We'll learn how to transform
- 3:21existing features and create new ones.
- 3:24Second, ensembling, combining multiple
- 3:26models to further boost uh predictive
- 3:29performance. Both techniques are
- 3:31directly useful for your competition
- 3:34score.
- 3:36Please download today's lecture
- 3:38materials from the students drive now.
- 3:40You will need You will need a notebook
- 3:42for the hands-on exercise after the
- 3:44video.
- 3:46Now, let's watch the lecture video.
About this transcript
This page contains the full transcript of GCI World 2026 September Session8 Opening by GCI World (Matsuo-Iwasawa Lab), generated from the public captions YouTube serves with the video. The transcript has 484 words across 92 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.