A blogpost from Apple summarizing their research on generative model for scene level radiance fields (GSN), ICCV 2021

https://machinelearning.apple.com/research/learning-to-generate-radiance-fields

👨‍🎓 Best Resources to Learn Natural Language Processing in 2021

https://www.kdnuggets.com/2021/09/best-resources-learn-natural-language-processing-2021.html

Anybody heard about semantic segmentation. Yeah, me too!

This article goes step by step into how semantic segmentation provides context to images through pixel-accuracy.

A very informative read that answers all the questions on the topic.
https://blog.superannotate.com/guide-to-semantic-segmentation/

Dual deployments on Vertex AI🔥🔥

Google introduced new deployment techniques for mobile and web.
https://cloud.google.com/blog/topics/developers-practitioners/dual-deployments-vertex-ai

DeepMind in collaboration with University College London released the "Reinforcement Learning Lecture Series 2021"

Website: https://deepmind.com/learning-resources/reinforcement-learning-series-2021
Lecture Video: https://youtube.com/playlist?list=PLqYmG7hTraZDVH599EItlEWsUOsJbAodm

Hello sir/ma'am,
I am Yashita Bawane, VII semester student of B. Arch from National Institute of Technology Raipur. I am working on my research paper titled "Enhancing productivity through daylighting in corporate workspaces".
As a part of my research work, I am undertaking a survey to know the lighting preferences of the corporate employees and how it relates to the quality of output.
I would be grateful if you spare some time and fill in the questionnaires. This survey would take 2-3 minutes to complete. Be assured that all the details will be kept confidential and will be used for academic purpose only.

Thank you in advance

https://forms.gle/5k46CdaFnG3HAJzw8

PySlowFast library by Facebook AI just added the neural Transformers support ☑☑

👉PySlowFast is an open source video understanding codebase from FAIR that provides state-of-the-art video classification models with efficient training

Highlights:
✅SlowFast Networks for Video Recognition
✅Supporting Non-local Neural Networks
✅Multigrid Method for Efficiently Training Video Models
✅X3D: Progressive Expansion for Efficient Video Rec.
✅Multiscale Vision Transformers (new!)

🚩Link to source code

Deploy your PyTorch model to Production

This tutorial aims to teach you how to deploy your recently trained model in PyTorch as an API using Python.

https://www.kdnuggets.com/2019/03/deploy-pytorch-model-production.html

Believe it or not, this is not originally a video. It was made from a collection of photos. Sounds interesting?

Learn more about video from this paper 👇

AI Synthesizes Smooth Videos from a Couple of Images!
Paper link: https://arxiv.org/pdf/2110.06635.pdf.
Code: https://github.com/darglein/ADOP

Here is a great tool that lets you visualize every single layer of 13 popular computer vision models such as ResNet, VGG, Inception v1 and v3, and CLIP.

The tool is Microscope by OpenAI
https://microscope.openai.com/models

International Conference on 3D Vision: 3DV 2021

Free Registration Deadline: 18th November

https://3dv2021.surrey.ac.uk/
Conference: 30th November - 3rd December 2021
Contact: a.hilton@surrey.ac.uk

Adjust your mindset for Machine Learning with Mark Ryan (Google Manager) 🚀

Get a chance to WIN free copies of Deep Learning with Structured Data book worth $35.99!!

To enter, share the LinkedIn post or Just comment your favorite part from this interview, Or you can also Retweet this tweet or just share your favorite part from this interview and tag us on Twitter.

Watch Podcast: https://youtu.be/iKPWTRhSJ4o?t=102

Google engineers offered 28 actionable tests for #machinelearning systems. 👇

Introducing 👉 The ML Test Score: A Rubric for ML Production Readiness and Technical Debt Reduction (2017). 👈

If #ml #training is like compilation, then ML testing shall be applied to both #data and code.

7 model tests

1⃣ 👉 Review model specs and version-control it. It makes training auditable and improve reproducibility.

2⃣ 👉 Ensure model loss is correlated with user engagement.

3⃣ 👉 Tune all hyperparameters. Grid search, Bayesian method whatever you use, tune all of them.

4⃣ 👉 Measure the impact of model staleness. The age-versus-quality curve shows what amount of staleness is tolerable.

5⃣ 👉 Test against a simpler model regularly to confirm the benefit more sophisticated techniques.

6⃣ 👉 Check the model quality is good across different data segment, e.g. user countries, movie genre etc.

7⃣ 👉 Test model inclusion by checking against the protected dimensions or enrich under-represented categories.

7 data tests

1⃣ 👉 Capture feature expectations in schema using statistics from data + domain knowledge + expectations.

2⃣ 👉 Use beneficial features only, e.g. training a set of models each with one feature removed.

3⃣ 👉 Avoid costly features. Cost includes running time, RAM as well as upstream work and instability.

4⃣ 👉 Adhere to feature requirements. If certain features can’t be used, enforce it programmatically.

5⃣ 👉 Set privacy controls. Budget enough time for new feature that depends on sensitive data.

6⃣ 👉 Add new features quickly. If conflicting with 5⃣ , privacy goes first.

7⃣ 👉 Test code for all input features. Bugs do exist in feature creation code.

See 7 Infrastructure & 7 monitoring tests in paper. 👇

They interviewed 36 teams across Google and found

👉 Using a checklist helps avoid mistakes (like a surgeon would do).

👉 Data dependencies leads to outsourcing responsibility. Other teams’ validation may not validate your use case.

👉 A good framework promotes integration test which is not well adopted.

👉 Assess the assessment to better assess your system.
https://research.google.com/pubs/archive/aad9f93b86b7addfea4c419b9100c6cdd26cacea.pdf

Advanced Computer Vision
Very Interesting course on Advance computer vision covers Dallie, Diffusion Models.
https://www.crcv.ucf.edu/courses/cap6412-spring-2023/schedule/
https://www.youtube.com/playlist?list=PLd3hlSJsX_In7qup928HaHmilugBGctuF