deep

manifesting an internship by the end of 2nd year

everytime i come home I'm greeted with few days of compulsary cold and fever 😷🏳️

Unpaid internships.

@sharpeye_wnl knock knock

trying to confirm if am shadow banned or not, if you see this say hi

@Zyara_1ot @knightkun__ sleep >_<

@deephivex @knightkun__

some mutual is talking like ruben veidt nowadays

@knightkun__ @Zyara_1ot 🤣😭

@Zyara_1ot @deephivex no gng it's the most ruben thing again 😭😭🥀🥀 dont

@Zyara_1ot @knightkun__ yeah yeah similar

@knightkun__ @deephivex ITS TIME TO DEACTIVATE X

@cneuralnetwork and bf was at iisc ( best physicist? / next nobel prize winner type somehting)

idk if you remember but there used to be a girl who worked in adobe research & iisc & bigbasket in her first year in 2023/24 times randomly yaad aagya

@Zyara_1ot @knightkun__ "nigga"

@deephivex @knightkun__ niggaaaaaaa deleteeeee thatttt poasttttttttttttt

@jerkeyray true 😆

@deephivex same but that's exactly why it's effective 😭

interview strat: ask the dude to open his chatgpt history and go through how does he use ai for his work

@Zyara_1ot @knightkun__ u seem very popular now

@knightkun__ @deephivex I DON'T TALK LIKE HIM OR LIKE ANYONE ELSE I TALK LIKE "ME" 😭OKEY

@Zyara_1ot @drrrrrrrr186 it's ok baccha calm down

@drrrrrrrr186 @deephivex HOW ME how?? NOU?? I DIDN'T SAY ANYTHING... LIKE ANYTHING

@drrrrrrrr186 @Zyara_1ot bwahahahahah

@Zyara_1ot @deephivex **and you have already been baited**

@aetosdios_ guess xD

@deephivex who

@knightkun__ xD

@deephivex did this mutual post an article today

@jerkeyray who's ur guess lol

@deephivex who???

@Zyara_1ot u poured yourself in the blog aah great one till date :⁠^⁠)

so, in practice, CNNs often combine both invariance and equivariance by using convolution layers to maintain shift-equivariance and pooling or striding layers to approximate shift invariance. thanks if u read this far :)

to infer relationships b/w feature maps that differ only by rotation. researchers have explored methods to mitigate this, such as using spherical harmonics to describe features in a rotation-invariant way, which helps the network to maintain stability despite i/p rotations

the central agenda and question to understand is to understand why CNNs are so effective at processing challenging data. this brings us to invariance and equivariance, which are essential for explaining how CNNs handle transformations like shifts in input data.