@deklili.ammora:

deklili.ammora
deklili.ammora
Open In TikTok:
Region: ID
Sunday 08 March 2026 10:41:05 GMT
19456
2003
11
156

Music

Download

Comments

nadyajessicaaaa
Gek nad :
yg tengah 💕
2026-03-08 20:17:07
0
arkaplat_p
Arka_plat p :
kerih kerih💃💃🤭
2026-03-14 15:27:05
0
rino.no51
RINOXS🀄🐲 :
okonge
2026-03-08 20:38:45
0
agunklonthongsuwu
agunklonthongsuwu :
kalem mbk lili 🥰🥰🥰
2026-03-08 20:21:07
0
dominic_miami
dominic1812 :
Yg tengah punya Om ini jangan di ambil ya🥰🫶🏻
2026-03-17 07:06:50
0
kaptainnaffi
kaptainnaffi :
🥰👌
2026-03-08 13:19:34
0
asfar_c772
ɪ’ᴍ ᴀ.ᴍ. :
🔥🔥🔥
2026-03-08 13:07:25
0
aden.teha
Aden teha :
🥰🥰🥰🥰🥰🥰
2026-03-08 11:30:23
0
janpait530
𝙲𝙰𝚁𝙻𝙾𝚂 🇲🇨🇨🇳 :
yg tengah jgn smpe Lepas
2026-03-08 22:00:31
0
To see more videos from user @deklili.ammora, please go to the Tikwm homepage.

Other Videos

**🤯 What if the
**🤯 What if the "layers" in Deep Learning are just an illusion?** Deep learning engineers originally built ResNets as a clever hack to fix vanishing gradients. But without realizing it, they had accidentally reinvented a 300-year-old mathematical concept: Euler’s Method for numerical integration. When you take that math and shrink the step size to zero, the discrete layers completely dissolve. The network becomes a continuous Ordinary Differential Equation (ODE). 🧠 **The Quick-Win Mental Shortcut:** Next time you read about Neural ODEs, just remember the "Assembly Line vs. Force Field" rule: • **Standard Network = The Assembly Line.** Data teleports between fixed stations (layers) and gets abruptly changed. • **Neural ODE = The Force Field.** There are no stations. The network just learns the "wind," and a mathematical solver continuously flows your data down the river. 🔗 **Want the full academic proof?** If you want to see the exact calculus limit that makes layers disappear, I just published a new Deep-Dive on Substack. We cover the step-by-step derivation, how the Adjoint Method allows training with $O(1)$ memory cost, and why this is the ultimate tool for messy time-series data. Link in bio! 💬 **Question for you:** Which mental model feels more intuitive to you when thinking about AI—the discrete assembly line or the continuous force field? Let me know in the comments! 👇 #machinelearning #deeplearning #neuralnetworks #calculus #differentialequations

About