Craig Brozefsky
Bartek JasickiAnd weekly wakie with the new development report from my #opensource projects. The list is a bit longer this week: a roguelike game, #Nim binding to Nuklear GUI library and static code analyzer for Nim were on the table. More info: www.laeran.pl.eu.org/blog/devblog...
Happy weekend everyone. ๐
@simon_brooke #nim perhaps? Your lispy predilections may be piqued, provided the Python whitespace perversions donโt piss you off.
Another week ends, another weekly development report from my #opensource projects is ready to read. As usual, a roguelike game and #Nim binding to Nuklear GUI library got my attention this week. More info here: www.laeran.pl.eu.org/blog/devblog...
As always, happy weekend everyone. ๐
Ars Technica NewsFiguring out why AIs get flummoxed by some games https://arstechni.ca/hwsf #Computerscience #training #Science #chess #games #math #NIM #AI
michael13-Mar-2026
#AIโs #gamePlaying still has flaws: #AlphaZero-style self-play tested on #Nim
Despite heavy training, agents show blind spots and can miss optimal moves
AIโs game-playing still has flaws: AlphaZero-style self-play tested on Nim
Games are often called the โFormula 1โ of AI: clear rules, clear winners. AlphaZero-style algorithms learn by self-play: a neural network predicts moves and guides tree search. We tested this recipe on Nim, a simple childrenโs matchstick game that has been mathematically solved. Because the correct move is known for every position, we can measure whether an agent plays optimally across the state space. We find a gap: learning can work on small boards, but blind spots remain and performance degrades as the board grows, with predictions approaching random. This suggests impartial games often need analytic representations, not pattern learning.
sayzardNVIDIA AI Developer (@NVIDIAAIDev)
Nemotron 3 Super๊ฐ ์ํฌ์คํ ์ด์ ๋ถํฐ ํด๋ผ์ฐ๋๊น์ง ๋ค์ํ ํ๊ฒฝ์ ๋ฐฐํฌ ์ง์๋๋ฉฐ API, OpenRouter, build.nvidia.com ๊ฒฝ์ ๋ก ์ ๊ทผ ๊ฐ๋ฅํ๋ค๋ ๊ณต์ง์ ๋๋ค. ์ฃผ์ ์ถ๋ก (inference) ํ๋ซํผ์์ NVIDIA NIM ํจํค์ง๋ก ์ ๊ณต๋์ด ๋ฐฐํฌ์ ์ฌ์ฉ์ด ์ฉ์ดํด์ก์ต๋๋ค.
NVIDIA AI Developer (@NVIDIAAIDev) on X
@openclaw @perplexity_ai Ready to get started? Nemotron 3 Super supports deployment across environments, from workstations to the cloud, and can be accessed through API, OpenRouter, or https://t.co/fC1rz1G9c4. It is now live and available on major inference platforms, packaged as NVIDIA NIM: ๐ฅ
Another Saturday, another weekly development report from my #opensource projects is ready to read. As almost usual, a roguelike game and a #Nim binding to Nuklear GUI library were on the table. More info: www.laeran.pl.eu.org/blog/devblog...
Happy weekend, everyone. ๐
cmsThe thing i really like about Nim is that whenever I think of a convenient language affordance, that's very helpful for some particular thing, it turns out they have one of those already! (Today I discovered they have source filters - https://nim-lang.org/docs/filters.html)
#nim #programming
#nim #programming
Lรช Fรช is backThis has improved estimation times dramatically for complex #SEM estimations. I wonder how much it can speed up #genomicSEM analyses. https://fastsem.netlify.app #nim
New Saturday, new weekly, development report from my #opensource projects is ready. As almost always, a roguelike game and #Nim binding to Nuklear GUI library were on the table. More information: www.laeran.pl.eu.org/blog/devblog...
As always, happy weekend, everyone. ๐