LinkedIn's Sakshi Jain gave an enlightening talk on fairness in machine learning. If you missed it, visit the link below to learn how the AI team at LinkedIn isworks to improve algorithmic fairness.
MACHINE LEARNING
-
Top 10 Machine Learning Algorithms for Digital Transformation
By
–
DIGITAL TRANSFORMATION
Top 10 Machine Learning Algorithms https://
bit.ly/3KkK0pK By @ingliguori @Kenovy_it #DigitalTransformation ##MachineLearning -
AlphaTensor: Novel Matrix Multiplication Discovery and Security Automation
By
–
That said, here's an example of inventing new matrix multiplication (do your homework): https://
deepmind.google/discover/blog/
discovering-novel-algorithms-with-alphatensor/
… Plus, security vulnerabilities are found routinely (and by default) with automated tools on all serious projects. -
AI Code Golf: Refactoring and Bug Detection in ML
By
–
Most significant is the API-compliance "code golf" type of code, which is so prevalent in ML. It's able to handle refactoring much more reliably, and helps find tricky bugs as a first-pass sanity check… I also let it write first pass of most new scripts, but they change a lot.
-
LLM2LLM: Iterative Data Augmentation Strategy for Model Fine-tuning
By
–
10). LLM2LLM – an iterative data augmentation strategy that leverages a teacher LLM to enhance a small seed dataset by augmenting additional data that can be used to effectively fine-tune models.
-
FollowIR Dataset Improves Information Retrieval Model Performance
By
–
9). FollowIR – a dataset with instruction evaluation benchmark and a separate set for teaching information retrieval model to follow real-world instructions; a FollowIR-7B model has significant improvements (over 13%) after fine-tuning on a training set.
-
Top ML Papers of the Week: DBRX, Grok, LLM2LLM and More
By
–
The Top ML Papers of the Week (March 25 – March 31): – DBRX
– Grok-1.5
– LLM2LLM
– Mini-Gemini
– Agent Lumos
– Long-form factuality in LLMs
… -
DBRX: New 132B Open-Source LLM Outperforms Benchmarks
By
–
1). DBRX – a new 132B parameter open LLM that outperforms all the established open-source models on common benchmarks like MMLU and GSM8K; DBRX was pretrained on 12T tokens (text and code) and uses a mixture-of-experts (MoE) architecture.
-
AI Recognized as Top Industry Disruptor by CEOs
By
–
#AI identified as top industry disruptor by CEOs
by @AaronPeterHurst @InformationAge Read more: https://
buff.ly/3opJQFD #BigData #MachineLearning #ArtificialIntelligence #ML #MI cc: @pascal_bornet @pbalakrishnarao @rtehrani -
Google’s New Robot Learns to Follow User Commands
By
–
Google's new #Robot has taught itself to follow your command
— Ronald van Loon (@Ronald_vanLoon) 31 mars 2024
by @Wef#Robotics #RPA #Automation #AI #MachineLearning #ArtificialIntelligence #ML #MI
cc: @dirkschaar @space_mog @make @maxjcm pic.twitter.com/bVmYpXqZufGoogle's new #Robot has taught itself to follow your command
by @Wef #Robotics #RPA #Automation #AI #MachineLearning #ArtificialIntelligence #ML #MI cc: @dirkschaar @space_mog @make @maxjcm