jin @_jinyeom

software engineer @microsoft Seattle, WA Joined August 2010

Tweets

5K
Followers

204
Following

730
Likes

6K

Yann LeCun @ylecun

2 years ago

I-JEPA: Efficient method for Self-Supervised Learning of image features. No need for data augmentation, just masking. Joint embedding predictive architecture, not generative. And it's open source, of course. Blog: ai.facebook.com/blog/yann-lecu… Paper: arxiv.org/abs/2301.08243 Code &…

AI at Meta @AIatMeta

2 years ago

37 335 2K 746K 524

50 306 2K 464K 526

Joel Lehman @joelbot3000

3 years ago

"Machine Love" - new paper w/ LM expts: Leaving to humans what is human (emotions/relationships/autonomy), is there any conception of love fitting for machines to embody? One idea: to provide unconditional support for humans to pursue their own growth. arxiv.org/pdf/2302.09248…

13 34 197 56K 58

Download Video

Luiz Pessoa @PessoaBrain

3 years ago

𝗥𝗲𝘄𝗮𝗿𝗱 𝗶𝘀 𝗻𝗼𝘁 𝗡𝗲𝗰𝗲𝘀𝘀𝗮𝗿𝘆: How to Create a Compositional Self-Preserving Agent for Life-Long Learning Absolutely tantalizing by @no_reward_for_u arxiv.org/abs/2211.10851

11 76 407 79K 181

Download Image

flavio @flaviocopes

3 years ago

Web3 devs turning into AI devs

88 1K 11K 975K 294

Download Image

Viacheslav Biriukov @brk0v

3 years ago

mmap() is a tremendously important Linux syscall. Many open-source and in-house databases and KV storages use it internally to simplify data access. That's why it's so essential for SREs to understand its interaction with Page Cache: biriukov.dev/docs/page-cach…

21 439 2K 336K 925

Download Image

jin @_jinyeom

3 years ago

Arch @OsEndeavour

1 0 0 101 0

Download Image

Tim Connors @itstimconnors

3 years ago

This fucking guy

726 6K 54K 0 3K

Download Image

Kevin Frans @kvfrans

3 years ago

Powderworld is a lightweight simulation environment for understanding AI generalization, supporting: - modular & emergent ruleset - expressive RL task space - 10k timesteps/sec via GPU Paper: arxiv.org/abs/2211.13051 Blog + Online Demo: kvfrans.com/static/powder/

14 114 640 0 203

Download Video

Kevin Frans @kvfrans

3 years ago

Btw, a fully-featured simulation of Powderworld runs in the browser. Since elements all interact locally, they can be implemented as Pytorch convolutional operations. We can compile this network into an ONNX function that runs via webGL. kvfrans.com/static/powder/

2 24 183 0 33

Download Video

Kenneth Stanley @kenneth0stanley

3 years ago

Great work @carperai ! Evolution through Large Models (ELM) in open source!

1 7 57 0 7

Farid @ai_fast_track

3 years ago

✨ Diffusion meets Object Detection 📰 DiffusionDet: Diffusion Model for Object Detection • It formulates object detection as a denoising diffusion process from noisy boxes to object boxes • At training, object boxes diffuse from ground-truth boxes to random distribution

13 125 611 0 244

Download Image

Samuel Arbesman @arbesman

3 years ago

Large-scale simulation is all around, from weather simulations to modeling mobility within cities. 🖥️🌍 But I still think that the creation of simulations needs to be democratized. In other words, where’s my “HyperCard for Simulation”? 🧵 1/

10 64 289 0 151

Download Image

trash @trashh_dev

3 years ago

hey. we like it, okay?

27 29 359 0 2

Download Image

Renjie Liao @lrjconan

3 years ago

We show that Gaussian RBMs can generate good images just like other generative models, despite the single-layer architecture. Key innovations: 1) Gibbs-Langevin sampling; 2) modified Contrastive Divergence. Paper: arxiv.org/abs/2210.10318 Code: github.com/lrjconan/GRBM 1/2

10 109 536 0 191

Download Gif

Jianwei Yang @jw2yang4ai

3 years ago

Excited to share our FocalNets achieve new SoTA 64.3(test)/64.2(val) on COCO object detection, w/ 3x smaller model/data size than prior arts! It is the first ATTENTION-FREE arch ever in the last 2yrs beating ALL attention-based models on this leaderboard! github.com/microsoft/Foca…

AK @_akhaliq

4 years ago

1 26 134 0 27

Download Image

8 69 410 0 134

Download Image

AK @_akhaliq

3 years ago

Cyberpunk-Anime-Diffusion: finetuned Waifu Diffusion V1.3 Model with Stable Diffusion V1.5 New Vae, training in Dreambooth by @DGSpitzer @huggingface model: huggingface.co/DGSpitzer/Cybe…