#ml

Machine Learning Visualized — Machine Learning Visualized
https://ml-visualized.com/?utm_source=substack&utm_medium=email

#ml

What’s Really Going On in Machine Learning? Some Minimal Models—Stephen Wolfram Writings
https://writings.stephenwolfram.com/2024/08/whats-really-going-on-in-machine-learning-some-minimal-models/

Stephenwolfram

What’s Really Going On in Machine Learning? Some Minimal Models

Stephen Wolfram explores minimal models and their visualizations, aiming to explain the underneath functionality of neural nets and ultimately machine learning.

#ml

Meta's second version of segment anything.

https://github.com/facebookresearch/segment-anything-2


They have a nice demo:

https://sam2.metademolab.com/

GitHub

GitHub - facebookresearch/sam2: The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM…

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th...

#ml

I was searching for a tool to visualize computational graphs and ran into this preprint. The hierarchical visualization idea is quite nice.

https://arxiv.org/abs/2212.10774

arXiv.org

Towards Efficient Visual Simplification of Computational Graphs in...

A computational graph in a deep neural network (DNN) denotes a specific data flow diagram (DFD) composed of many tensors and operators. Existing toolkits for visualizing computational graphs are...

#ml

Schmidhuber J. Deep Learning: Our Miraculous Year 1990-1991. In: arXiv.org [Internet]. 12 May 2020 [cited 7 Jul 2024]. Available: https://arxiv.org/abs/2005.05744

arXiv.org

Deep Learning: Our Miraculous Year 1990-1991

In 2020-2021, we celebrated that many of the basic ideas behind the deep learning revolution were published three decades ago within fewer than 12 months in our "Annus Mirabilis" or "Miraculous...

#ml

https://github.com/google-research/timesfm

GitHub

GitHub - google-research/timesfm: TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed…

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting. - google-research/timesfm

#ml

Like a dictionary


Kunc, Vladim’ir, and Jivr’i Kl’ema. 2024. “Three Decades of Activations: A Comprehensive Survey of 400 Activation Functions for Neural Networks.” arXiv [Cs.LG], February. http://arxiv.org/abs/2402.09092.

arXiv.org

Three Decades of Activations: A Comprehensive Survey of 400...

Neural networks have proven to be a highly effective tool for solving complex problems in many areas of life. Recently, their importance and practical usability have further been reinforced with...

#ml

I got interested in satellite data last year and played with it a bit. It's fantastic. The spatiotemporal nature of it brings up a lot of interesting questions.

Then I saw this paper today:


Rolf, Esther, Konstantin Klemmer, Caleb Robinson, and Hannah Kerner. 2024. “Mission Critical -- Satellite Data Is a Distinct Modality in Machine Learning.” arXiv [Cs.LG], February. http://arxiv.org/abs/2402.01444.

arXiv.org

Mission Critical -- Satellite Data is a Distinct Modality in...

Satellite data has the potential to inspire a seismic shift for machine learning -- one in which we rethink existing practices designed for traditional data modalities. As machine learning for...

#ml

Jelassi S, Brandfonbrener D, Kakade SM, Malach E. Repeat after me: Transformers are better than state space models at copying. arXiv [cs.LG]. 2024. Available: http://arxiv.org/abs/2402.01032

Not surprising at all when you have direct access to a long context. But hey, look at this title.

arXiv.org

Repeat After Me: Transformers are Better than State Space Models at Copying

Transformers are the dominant architecture for sequence modeling, but there is growing interest in models that use a fixed-size latent state that does not depend on the sequence length, which we...

#ml


Interesting idea to use Hydra in ML experiments.

https://github.com/ashleve/lightning-hydra-template

GitHub

GitHub - ashleve/lightning-hydra-template: PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡

PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡ - ashleve/lightning-hydra-template

#ml

Hand-Crafted Transformers


HandCrafted.ipynb - Colaboratory
https://colab.research.google.com/github/newhouseb/handcrafted/blob/main/HandCrafted.ipynb

Google

HandCrafted.ipynb

Run, share, and edit Python notebooks

#ml

A family tree shows how transformers are evolving.

(HTML is probably the worst name for a model.)

https://arxiv.org/abs/2302.07730

#ml

Yes, Transformers are Effective for Time Series Forecasting (+ Autoformer)
https://huggingface.co/blog/autoformer

huggingface.co

Yes, Transformers are Effective for Time Series Forecasting (+ Autoformer)

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

#ml

https://www.kdnuggets.com/2023/06/ten-years-ai-review.html

KDnuggets

Ten Years of AI in Review - KDnuggets

From image classification to chatbot therapy.

#ml

Yeh, Catherine, Yida Chen, Aoyu Wu, Cynthia Chen, Fernanda Viégas, and Martin Wattenberg. 2023. “AttentionViz: A Global View of Transformer Attention.” ArXiv [Cs.HC]. arXiv. http://arxiv.org/abs/2305.03210.

#ml

Pérez J, Barceló P, Marinkovic J. Attention is Turing-Complete. J Mach Learn Res. 2021;22: 1–35. Available: https://jmlr.org/papers/v22/20-302.html

#ml

https://mlcontests.com/state-of-competitive-machine-learning-2022/

Quote from the report:

Successful competitors have mostly converged on a common set of tools — Python, PyData, PyTorch, and gradient-boosted decision trees.

Deep learning still has not replaced gradient-boosted decision trees when it comes to tabular data, though it does often seem to add value when ensembled with boosting methods.
Transformers continue to dominate in NLP, and start to compete with convolutional neural nets in computer vision.

Competitions cover a broad range of research areas including computer vision, NLP, tabular data, robotics, time-series analysis, and many others.
Large ensembles remain common among winners, though single-model solutions do win too.

There are several active machine learning competition platforms, as well as dozens of purpose-built websites for individual competitions.
Competitive machine learning continues to grow in popularity, including in academia.

Around 50% of winners are solo winners; 50% of winners are first-time winners; 30% have won more than once before.

Some competitors are able to invest significantly into hardware used to train their solutions, though others who use free hardware like Google Colab are also still able to win competitions.

ML Contests

The State of Competitive Machine Learning | ML Contests

We summarise the state of the competitive landscape and analyse the 200+ competitions that took place in 2022. Plus a deep dive analysis of 67 winning solutions to figure out the best strategies to win at competitive ML.

#ml

Releasing the Skynet

https://github.com/internet-explorer-ssl/internet-explorer

GitHub

GitHub - internet-explorer-ssl/internet-explorer: Internet Explorer explores the web in a self-supervised manner to progressively…

Internet Explorer explores the web in a self-supervised manner to progressively find relevant examples that improve performance on a desired target dataset. - GitHub - internet-explorer-ssl/interne...

#ml

google-research/tuning_playbook: A playbook for systematically maximizing the performance of deep learning models.
https://github.com/google-research/tuning_playbook

GitHub

GitHub - google-research/tuning_playbook: A playbook for systematically maximizing the performance of deep learning models.

A playbook for systematically maximizing the performance of deep learning models. - google-research/tuning_playbook