DKM: Differentiable K-Means Clustering Layer for Neural Network Compression deepai.org/publication/dk… by Minsik Cho et al. including Mohammad Rastegari #KMeans #NaturalLanguageProcessing

Here is our work at Apple on building a state-of-the-art parameter efficient visual transformer. @apple_ml_research

Deploying ML models on device cannot be static. ML models should adapt themselves to available resource. In our recent research in #Apple, we learn ML models that can dynamically be compressed to any arbitrary sparsity or quantized level at inference time.

Yes, we are releasing code in Apple to promote effective contribution in our research community. #apple_ml_research, #apple, #DeepLearning

Wow deep attacks seems to be a very serious problem in our ML modeling approach. Perhaps something is fundamentally wrong in our models!!! arxiv.org/abs/1910.00744 #deeplearning

Great progress in self-supervised training. Next milestone matching supervised SOAT Imagenet1k with self-supervised without end-to-end fine tuning

These CVPR policies are frustrating. Given all the randomness in the review process I feel there is no point submitting paper into conferences anymore. By the law of large data (large number of papers and readers) just submitting to arXiv will be enough for a good paper to shine.

A Transformer that can be as efficient as a CNN but yet maintains the high performance in the large data regime training. #EfficientTransformer, #AppleMI

We provide an empirical analysis over different sharing strategies in isotropic networks and how they can make large networks memory-efficient. Joint work with Chien-Yu Lin, Anish Prabhu, Thomas Merth, Sachin, Maxwell Horton, Mohammad Rastegari. Code: github.com/apple/ml-spin.

MobileOne code is released.

Introducing NeuMan, a NeRF representation of human together with the scene. From a single clip (<100 frames), NeuMan can perform view synthesis of the scene without/with the human in novel poses. (1/4) project page: machinelearning.apple.com/research/neura…… code: github.com/apple/ml-neuman

NeuMan is a new #ECCV2022 paper from our research team Apple. Using a short (~10s) clip, we reconstruct human and scene radiance fields, and re-render with novel human poses and views. Paper/code/videos: machinelearning.apple.com/research/neura… w. Wei Jiang, G. Samei, Kwang Moo Yi, Anurag Ranjan

“Reinforce Data, Multiply Impact: Improved Model Accuracy and Robustness with Dataset Reinforcement” is an #iccv2023 paper from #Apple. By just swapping the ImageNet dataset with the “reinforced” ImageNet+ dataset, a model can be trained up to 7x faster to reach the same accuracy

Accurate training aware weight quantization was computationally intractable for LLMs. But now in Apple MIND we developed a method to solve the problem very efficiently and it pushes the boundary to 3-bit quantization. eDKM: arxiv.org/abs/2309.00964 #LLM, #LLMoptimizaton

This has been one of my favorite directions on enabling #llms to run effectively on device. Thanks to the great team who are pushing state-of-the-art in this direction. In the Apple MIND team, we try to attack research problems that move us to the next level of experiencing AI.

Glad to see a great attention to our research in the Apple MIND team.

Apple presents Speculative Streaming Fast LLM Inference without Auxiliary Models Speculative decoding is a prominent technique to speed up the inference of a large target language model based on predictions of an auxiliary draft model. While effective, in application-specific

Another great research effort on efficiency of LLMs from our group in #apple

This work was one of the last works that was done by my team when I was working at Apple. A lot of credit to Sachin whose dedication was the key to this project. Main point behind here is to show as a contributor to the AI community we play our role to be fully open.

Like OpenELM, CatLIP is also "Open" github.com/apple/corenet