La Guerre des Intelligences a commencé
MACHINE LEARNING
-
Testing embedding strategy with subword features integration
By
–
Okay so yeah they do a variety of things and their experiments seem to make sense. I'd like to try plugging our embedding strategy into their thing. The subword features are pretty helpful imo.
-
Resolution Life Partners with H2O AI for Claims Segmentation
By
–
Australasia's largest life insurer, #ResolutionLife, has partnered with @h2oai to improve: #CustomerService by segmenting claims in 15 seconds Model accuracy by 6% Check out this video with Rajesh Malla (Head of Data Engineering). #machinelearning https://
youtube.com/watch?v=d6Ev3T
GWMLk
… -
Curriculum Learning with Positive Examples: Don’t Shoot the Dog
By
–
Have you read Don't Shoot the Dog? I found it super interesting, there's the idea of curriculum learning with only positive examples first then negative examples for refinement https://
amazon.ca/Dont-Shoot-Dog
-Teaching-Training/dp/0553380397
… -
Research sandbox for measuring functional correctness of generated code
By
–
In research contexts, inside a sandbox, to measure functional correctness of generated code e.g. in the original Codex paper: https://
arxiv.org/abs/2107.03374 -
Reductionist Explanations Fail to Explain Complex Phenomena
By
–
“Particle collisions,” “chemical imbalance,” and “microbiome” are terrible explanations. Explanations that explain everything explain nothing.
-
Meta AI MCC Project: Paper and Project Page
By
–
Great examples in the project page: Meta AI MCC Project Authors: Chao-Yuan Wu, Justin Johnson, Jitendra Malik, Christoph Feichntenhofer, Georgia Gkioxari Paper: https://
arxiv.org/pdf/2301.08247
.pdf
… Project page: https://
mcc3d.github.io -
MCC recognizes 3D objects from single image using RGB-D video library
By
–
With Multiview Compressive Coding (MCC), the model tries to recognize 3D objects from a single image by pulling from large amounts of data from a library of RGB-D videos (recordings that capture depth information) to compare it to. pic.twitter.com/2Jte4QLp04
— AI Breakfast (@AiBreakfast) 29 janvier 2023With Multiview Compressive Coding (MCC), the model tries to recognize 3D objects from a single image by pulling from large amounts of data from a library of RGB-D videos (recordings that capture depth information) to compare it to.
-
Meta AI creates 3D models from a single image
By
–
Meta AI is working on creating 3D models from a single image, whether it's real or AI generated: pic.twitter.com/CJ0UjQvkgu
— AI Breakfast (@AiBreakfast) 29 janvier 2023Meta AI is working on creating 3D models from a single image, whether it's real or AI generated:
-
Generative Design: Where Nature, Innovation, and Technology Converge
By
–
How #Nature #inspiraton& #tech combine!
Explore #innovation in #Generative #Design Advances: #creativity #Sustainability & more! @Autodesk_UKI http://
bit.ly/GenDesign1 #AR #partner #100DaysOfCode #BIM #IoT #photo @BIM_PLUS #CAD @iangertler @mikeflache #AI @Shi4Tech #ML