Yunusa

Postdoc @UniOulu, Researcher #ComputerVision #DeepLearning #MedicalImageAnalysis #AI4Science

Lived in: 🇳🇬 🇨🇾 🇨🇳 🇫🇮

[Yunusa @_yunusa_]

@Alah8884 @Amfagge_4 Allah yasa ku sauke hakin mutane, expert.

[Yunusa @_yunusa_]

@Sarki_sultan @Abbacdo01 matsalar bin matan mutane fa?

[Yunusa @_yunusa_]

@Aabdul_HameedD @AbbaM_Abiyos Of course it's not, now you are making hypothetical arguments. Sexual coercion is not the solution, what about he raped her and she dies as a result?

[Yunusa @_yunusa_]

@AbbaM_Abiyos @Aabdul_HameedD 😁 shi kuma wanda yake neman matar aure shine mai ilimi? sexual coercion? It's so simple to think they should just report but reality is these same lecturers are the authorities. They are criminal syndicate preying on innocent vulnerable young ladies. Tirrr da irin wannan mutanen

[Yunusa @_yunusa_]

By incentivizing reproducibility, we can improve research quality, reduce wasted effort and help the community push the boundaries of AI faster and with greater confidence.

#happyweekends 😊

[Yunusa @_yunusa_]

A Call to take Reproducibility Seriously
Too often, reproducing a paper from top AI venues including #CVPR, #ICCV, ECCV, ICML, #NeurIPS, AAAI takes days or even months when it should take only a few hours.

[Yunusa @_yunusa_]

Broken repositories, missing code, undocumented dataset modifications inconsistencies between the paper and the released implementation slow scientific progress and make it unnecessarily difficult for researchers to build upon existing work.

[Yunusa @_yunusa_]

Rather than treating reproducibility as an afterthought, the community should reward it. Venues could introduce dedicated reproducibility tracks or awards and recognize high-quality reproduction studies as valuable scientific contributions.

[Kosta Derpanis (back home, Toronto ❤️) @CSProfKGD]

#KostasThoughts: The #computervision community is behind on review transparency. Reviews and meta-reviews should be made public for accepted papers and, if authors opt in, for rejected papers as well.

Transparency can be a powerful mechanism for improving review and meta-review quality.

2028 is coming soon 🤔😉

(Yeah, 😡 with one of our meta-reviews!)

[Yunusa @_yunusa_]

@CSProfKGD You said it all, God bless

[Yunusa @_yunusa_]

@eccvconf @eccvconf what are you guys cooking? half done is OK.

[Yunusa @_yunusa_]

spent the week training visual models. spent saturday meeting the original masters of vision and navigation. Owls and eagles don't need GPUs. 🦉🦅

[Yunusa @_yunusa_]

@gabriberton @georgiagkioxari @taiyasaki Thanks for sharing @gabriberton

[Yunusa @_yunusa_]

comfort zone 👽

[Yunusa @_yunusa_]

@jbhuang0604 Most young researchers are trained to think outside the box not to present beautifully. and for many, english isn't even their first language. Cut them some slack. Though the room is full of smart weirdos 😁. Go have some fun.

[Yunusa @_yunusa_]

@peter_richtarik @Dirque_L 😄

[Yunusa @_yunusa_]

@ylecun @CSProfKGD @oier_mees 😆😁

[Yunusa @_yunusa_]

@Son_Of_Herdsmen @BashirAhmaad 😁 kumafa hakane, tohm suyi hakuri