Sign in

Yuji Kanagawa

@kngwyu.bsky.social
45 followers 48 following 137 posts

Looking for job now ALIFE, neuroscience, running, and some random photos kngwyu.github.io

PostsRepliesMedia
Yuji Kanagawa @kngwyu.bsky.social · 10/10/2026
Suwako trail last week #photography
020
Yuji Kanagawa @kngwyu.bsky.social · 10/10/2026
Thanks! Well I'm now interning at a company and figuring out what I can do 😅
000
Yuji Kanagawa @kngwyu.bsky.social · 01/10/2026
Fellowship rejected 😓
110
Yuji Kanagawa @kngwyu.bsky.social · 28/09/2026
Finally I came back to mountain running #photography
020
Yuji Kanagawa @kngwyu.bsky.social · 23/09/2026
Vibe-coded a web app for checking irregular guitar tunings kngwyu.github.io/guitar-tunin... So far I'm only using it for Open G but maybe useful for sonic youth like weird tunings? I feel like I should make use of more ChatGPT tokens not to waste my money but I rarely want something new 🙄
kngwyu.github.io
Guitar Tuning Tool
Select a guitar tuning and highlight notes in your selected scale on the fretboard. Customize individual strings and note selections.
000
Yuji Kanagawa @kngwyu.bsky.social · 13/09/2026
chill #photography
030
Yuji Kanagawa @kngwyu.bsky.social · 04/09/2026
For me it took a while to admit that I'm interested in reinforcement learning but not that excited about robots (or... Physical AI?)
010
Yuji Kanagawa @kngwyu.bsky.social · 26/08/2026
From the Wikipedia article, the ancestor of whales lived in northern India. So why did they happen to live in the sea? 🤔
010
Yuji Kanagawa @kngwyu.bsky.social · 23/08/2026
If the proof in the appendix is 30 pages long you don't have to feel like you're cheating 😓
011
Yuji Kanagawa @kngwyu.bsky.social · 19/08/2026
Not as good as Okinawa but sunset still looks like sunset #photography
070
Yuji Kanagawa @kngwyu.bsky.social · 16/08/2026
My favorite running road got broken by heavy rain 😯
020
Yuji Kanagawa @kngwyu.bsky.social · 15/08/2026
Short paper I presented at GECCO 2026 "Evolution of Fear and Social Rewards in Prey-Predator Relationship" is now available on ACM dl.acm.org/doi/10.1145/...
dl.acm.org
Evolution of Fear and Social Rewards in Prey-Predator Relationship | Proceedings of the Genetic and Evolutionary Computation Conference Companion
You will be notified whenever a record that you have chosen has been cited.
030
Yuji Kanagawa @kngwyu.bsky.social · 13/08/2026
It's really tiring for me to continue applying for positions for which I will be selected with < 10 % prob... (optimistically says) I really want to come back to 'research' now 😭
120
Yuji Kanagawa @kngwyu.bsky.social · 31/07/2026
Last day in Okinawa #photography
070
Yuji Kanagawa @kngwyu.bsky.social · 22/07/2026
Sunset today #photography
040
Yuji Kanagawa @kngwyu.bsky.social · 21/07/2026
Heavy rain just came when we arrived at the crater of Poas volcano... but I could see it at least #photography
Poas volcano, Costa Rica
Hidden by crowds
030
Yuji Kanagawa @kngwyu.bsky.social · 16/07/2026
Presented my poster kngwyu.github.io/fear-prey-pr... at #GECCO2026 today. Thank you everyone for coming or at least taking a glance on it ♥
020
Yuji Kanagawa @kngwyu.bsky.social · 15/07/2026
Enjoying Costa Rica so far except that I understand only a few Spanish words #photography
050
Yuji Kanagawa @kngwyu.bsky.social · 09/07/2026
I'm a bit anxious because I've been obsessed with postdoc applications and done no research for a while 😓
010
Yuji Kanagawa @kngwyu.bsky.social · 07/07/2026
Sunset today #photography
040
Yuji Kanagawa @kngwyu.bsky.social · 20/06/2026
Sunset today #photography
030
Yuji Kanagawa @kngwyu.bsky.social · 12/06/2026
I mean, the methodology is quite different. I feel it's OK, but they don't.
010
Yuji Kanagawa @kngwyu.bsky.social · 12/06/2026
That's actually a difficult problem! I guess some reviewers in the complex systems field have a physics background, and I feel like they have a bit different research tastes than I do, because I'm from CS/AI 🤔 I understand the difference, but I'm not that good at convincing them
100
Yuji Kanagawa @kngwyu.bsky.social · 12/06/2026
I'm gradually understanding the benefit of evolutionary game theory, but still, I don't have a clever idea how to respond to a reviewer's comment like "Your evolutionary simulation is too original and does not have much connection to existing literature like evolutionary game theory" 🤔
230
Yuji Kanagawa @kngwyu.bsky.social · 04/06/2026
sunset yesterday #photography
040
Yuji Kanagawa @kngwyu.bsky.social · 31/05/2026
Also preparing for the typhoon I guess #photography
030
Yuji Kanagawa @kngwyu.bsky.social · 12/05/2026
Isn't it a good quote? > It (evolutionary psychiatry) instead provides a framework that can integrate knowledge from other basic sciences. From onlinelibrary.wiley.com/doi/full/10....
000
Yuji Kanagawa @kngwyu.bsky.social · 02/05/2026
Sunset today #photography
070
Yuji Kanagawa @kngwyu.bsky.social · 29/04/2026
Sunset yesterday #photography
050
Yuji Kanagawa @kngwyu.bsky.social · 22/04/2026
After 10 years, Bootstrapped DQN still provides fundamental insights into exploration to me 🤔 arxiv.org/abs/1602.04621
arxiv.org
Deep Exploration via Bootstrapped DQN
Efficient exploration in complex environments remains a major challenge for reinforcement learning. We propose bootstrapped DQN, a simple algorithm that explores in a computationally and statistically...
140
Yuji Kanagawa @kngwyu.bsky.social · 10/04/2026
Sunset today #photography
040
Yuji Kanagawa @kngwyu.bsky.social · 05/04/2026
Sunset today #photography
040
Yuji Kanagawa @kngwyu.bsky.social · 04/04/2026
So I went to watch a football match to recover from paper rejection and they lost 😓
020
Yuji Kanagawa @kngwyu.bsky.social · 03/04/2026
Thank you. I will try to calm down and keep myself
010
Yuji Kanagawa @kngwyu.bsky.social · 03/04/2026
Hmm I'm sorry to hear that. Hope you find out a good position eventually! BTW I'm not sure I have a goal now. I just wanted to know how we could replicate the process of brain evolution. I don't see any clear goal beyond journal submission tbh...
100
Yuji Kanagawa @kngwyu.bsky.social · 03/04/2026
So true 😭
110
Yuji Kanagawa @kngwyu.bsky.social · 03/04/2026
So I have achieved nothing throughout my 5 years PhD life 😢
110
Yuji Kanagawa @kngwyu.bsky.social · 03/04/2026
Maybe I should leave academia now 😓
010
Yuji Kanagawa @kngwyu.bsky.social · 03/04/2026
My submission to journal rejected... can I get PhD? 😱
010
Yuji Kanagawa @kngwyu.bsky.social · 02/04/2026
Thank you for your input! I'm doing a bit weird things (computational eco + evo + neuroscience...?) so I may be outsider in any community, but definitely interested in those communities and conferences
110
Yuji Kanagawa @kngwyu.bsky.social · 02/04/2026
I just noticed that only 11 'Complex Systems' papers got accepted to GECCO 2026 (and mine is one of these) Does this mean most researchers aim at ALIFE? 🤔 Or the world may not have many 'complex systems' researchers...
130
Yuji Kanagawa @kngwyu.bsky.social · 30/03/2026
Why is there no AI benchmark on Cataclysm: DDA? Or shoud I make it
000
Yuji Kanagawa @kngwyu.bsky.social · 29/03/2026
I'm really struggling to summarize my research plan to a 3-page postdoc fellowship application format 😭😭😭 Always feeling I'm missing something important...
110
Yuji Kanagawa @kngwyu.bsky.social · 28/03/2026
I'm still thinking about so many 'what if' scenarios (e.g., what if computer or robots can create new hardware) but I think I understand the point. Thank you for your explaination.
110
Yuji Kanagawa @kngwyu.bsky.social · 28/03/2026
Came too early for sunset #photography
ocean over forest
090
Yuji Kanagawa @kngwyu.bsky.social · 27/03/2026
I finally stopped worrying about my academic carrer because I believe I would have nothing to do but research after every possible job got replaced by AI. Then AI will assign research budget based on our proposal...... Not sure I should use AI to write the research proposal for AI though
110
Yuji Kanagawa @kngwyu.bsky.social · 27/03/2026
What do you think about the open-ended evolution of 'culture'? I talked with a language evolution researcher who told me that language is open ended because the combination can realise something that doesn't exist (e.g. 'white tomato'). Still the same agenda applies?
110
Yuji Kanagawa @kngwyu.bsky.social · 26/03/2026
I feel like if a 6D phase plot is possible, my evolutionary simulation results should be a lot easier to understand
020
Yuji Kanagawa @kngwyu.bsky.social · 25/03/2026
Too windy for BBQ #photography
040
Yuji Kanagawa @kngwyu.bsky.social · 25/03/2026
I see. I don't know anything about the company politics but those algorithms are interesting. I'm not sure I like the naming of Kondo gate though...
010