BayesianMonk's profile picture. Welcome to the Bayesian Monk!! This account is for connecting with fellow monks and learning, doing, teaching and meditating in the mystic universe of data.

BayesianMonk

@BayesianMonk

Welcome to the Bayesian Monk!! This account is for connecting with fellow monks and learning, doing, teaching and meditating in the mystic universe of data.

Pinned

Looks like Meta's LLaMa 65B parameter language model has reached AGI and showing human-like behavior. When you ask it to be nice, it doubles down on being a jerk. 🤷‍♂️ @ykilcher first raised alarms here: youtu.be/E5OnoYF2oAk?t=… @MetaAI #AGI #LLMs

BayesianMonk's tweet image. Looks like Meta's LLaMa 65B parameter language model has reached AGI and showing human-like behavior. When you ask it to be nice, it doubles down on being a jerk. 🤷‍♂️

@ykilcher first raised alarms here: youtu.be/E5OnoYF2oAk?t=…

@MetaAI 

#AGI  #LLMs

BayesianMonk reposted

A major engine of innovation and economic development is startups, or "Little Tech" as @pmarca calls it. Governments should do what they can to incentivize and facilitate the creation and growth of startups. But startups can thrive only if a number of conditions are present: 1-…

This post is unavailable.

BayesianMonk reposted

# on shortification of "learning" There are a lot of videos on YouTube/TikTok etc. that give the appearance of education, but if you look closely they are really just entertainment. This is very convenient for everyone involved : the people watching enjoy thinking they are…


BayesianMonk reposted

"Voyage through Time" is my first artpiece using #stablediffusion and I am blown away with the possibilities... We're crossing a threshold where generative AI is no longer just about novel aesthetics, but evolving into an amazing tool to build powerful, human-centered narratives


BayesianMonk reposted

Food for thought: When optimizing multiple objectives, instead of training on loss = α loss + β loss, it's easier to train 2 models & combine their scores score = α score + β score 💡 Makes the ML system easier to update and maintain; doesn't require retraining when α & β change


Truer words have never been spoken. @mrdbourke Check this out if you haven't: youtube.com/watch?v=fw6NMQ…

BayesianMonk's tweet image. Truer words have never been spoken. @mrdbourke Check this out if you haven't: youtube.com/watch?v=fw6NMQ…

Early distress signals! Transformer model thinks machines taking over is a positive. 😅 #MachineLearning #AI #Singularity

BayesianMonk's tweet image. Early distress signals! Transformer model thinks machines taking over is a positive. 😅 #MachineLearning #AI #Singularity

BayesianMonk reposted

When it comes to data science, the most underrated skill is communication. Even if your insights and models are great, it makes a huge difference how you communicate them within or outside your team.


Can't wait to read this one. Thanks @sirbayes #DataScience

I am delighted to announce that a draft of my latest book, “Probabilistic Machine Learning: Advanced Topics”, is now available online at probml.ai. It covers #DeepGenerativeModels, #BayesianInference, #Causality, #ReinforcementLearning, #DistributionShift, etc.

sirbayes's tweet image. I am delighted to announce that a draft of my latest  book, “Probabilistic Machine Learning: Advanced Topics”, is now available online at probml.ai. It covers #DeepGenerativeModels, #BayesianInference, #Causality, #ReinforcementLearning, #DistributionShift, etc.


BayesianMonk reposted

I'm thrilled to announce that we're releasing the fastest #autoarima implementation for #Python today! 😍 It is a mirror implementation of @robjhyndman's #autoarima (R) and it is optimized using @numba_jit. It is 20x faster than pmdarima and @MetaAI's Prophet. 😎 🧵 @nixtlainc


BayesianMonk reposted

🐙 Transformer Recipe 🐙 There's a lot of material on learning about Transformers so I've written a concise list of the most useful study materials I've come across. From high-level introductions to code implementations. github.com/dair-ai/Transf…

omarsar0's tweet image. 🐙 Transformer Recipe 🐙

There's a lot of material on learning about Transformers so I've written a concise list of the most useful study materials I've come across. From high-level introductions to code implementations.

github.com/dair-ai/Transf…

BayesianMonk reposted

Excited to share that our colab tutorial on graph neural networks is live on Github under deepmind/educational! It also introduces #jraph, @DeepMind's in-house graph library for the #JAX stack. Co-authored with @ni_jovanovic. dpmd.ai/gnn-jraph-colab

lisawang1010's tweet image. Excited to share that our colab tutorial on graph neural networks is live on Github under deepmind/educational! It also introduces #jraph, @DeepMind's in-house graph library for the #JAX stack. Co-authored with @ni_jovanovic.
dpmd.ai/gnn-jraph-colab

Singularity is here!! @DeepLearningAI_

Guess our work here is done.

DeepLearningAI's tweet image. Guess our work here is done.


Musing of the day: If you can crawl, walk. If you can walk, jog. If you can jog, run. Progressive overload will do wonders when learning a new skill. Data science is certainly one such skill where you learn more by doing! #DataScience #100DaysOfCode #GrowthMindset #Learn


BayesianMonk reposted

Graph neural networks (GNNs) are rapidly advancing progress in ML for complex graph data applications. Let's have a look at some resources to help you learn and keep up-to-date with GNNs ↓

omarsar0's tweet image. Graph neural networks (GNNs) are rapidly advancing progress in ML for complex graph data applications.

Let's have a look at some resources to help you learn and keep up-to-date with GNNs ↓

BayesianMonk reposted

"Speech and Language Processing" by @jurafsky and James H. Martin remains my go-to reference for all things natural language processing (NLP). web.stanford.edu/~jurafsky/slp3/ It just has the right structure to learn about NLP. It's regularly updated and publicly accessible.

omarsar0's tweet image. "Speech and Language Processing" by @jurafsky and James H. Martin remains my go-to reference for all things natural language processing (NLP). web.stanford.edu/~jurafsky/slp3/

It just has the right structure to learn about NLP. It's regularly updated and publicly accessible.

No. 3 for fellow data nerds. #Python #techtwitter #DataScience #100DaysOfCode

5 GitHub Repositories To Boost Your Python Programming 👇[Thread]



Loading...

Something went wrong.


Something went wrong.