1/Several leaders share their hopes for AI in 2023, including finding key missing pieces that will enable algorithms to reason, building a personal data timeline, improving AI processes, discovering new principles for explainability, and using generative AI for active learning.
MACHINE LEARNING
-
How AI learns animal-environment correlations in image generation
By
–
Fijaos un detalle muy curioso que enseña la forma de operar de la IA, cómo al pedir convertir al gato en conejo, toooodo el entorno de alrededor se convierte en indoor (fondo, puerta, suelo, etc) Esto es porque la IA ha aprendido cuál es la correlación animal-entorno habitual 🙂
-
Image Generation Model Evaluation: Quality Beyond FID Metrics
By
–
Son imágenes chulas (hace un año hubieran sido increíbles), pero me parecen planas y con texturas con muy poco detalle. Cuantitativamente no se aleja mucho de Imagen y supera a Stable Diffusion, pero no sé. ¿A lo mejor evaluar con la métrica FID se nos está quedando limitado?
-
Interest in Small Local Language Models Under 1 Billion Parameters
By
–
i’d LOVE a cheap small (sub 1B) model to run this stuff locally! actually am wondering if gpt2 does this well enough
-
AI Solves 10 International Math Olympiad Problems with HyperTree Method
By
–
Using our HyperTree Proof Search method we created a neural theorem solver that was able to solve 10 International Math Olympiad problems — 5x any previous AI system. More in this post ➡️ https://t.co/dec9VeJhny pic.twitter.com/rns9IffWqx
— AI at Meta (@AIatMeta) 3 janvier 2023Using our HyperTree Proof Search method we created a neural theorem solver that was able to solve 10 International Math Olympiad problems — 5x any previous AI system. More in this post https://
bit.ly/3vBnu3E -
MAML Meta-Learning and Software Architecture Design Principles
By
–
reminds me of MAML meta-learning (
https://
arxiv.org/abs/1703.03400) where the objective is to find weights of a network such that any new task finetunes fast. In Software 1.0 land, equivalent is writing code such that any new desired functionality is simple and doesn't need a refactor. -
2023 AI Predictions: GPU Demand Soars with Generative AI Growth
By
–
What does 2023 have in store for AI? Our CEO Nick Elprin predicts GPU demand will soar as more deep learning apps emerge to cure disease. @NVIDIAAI
's Manuvir Das concurs that new generative AI realities will do the same. @VentureBeat has the full takes: https://
domino.buzz/3CfEKz5 -
Tech entrepreneur discovers transformers excel at transformation tasks
By
–
haha this is a great point: https://
x.com/markopolojarvi
/status/1610272055097757697?s=20&t=SZBzJCje8k8mxsn7HcYZUQ
… This just in – techbro realizes that transformers are really good at transforming things. More at 11 -
YOLOv5 Instance Segmentation: Elevate Your Object Detection Projects
By
–
🚀Blog-tastic Tuesdays🚀
— Satya Mallick (@LearnOpenCV) 3 janvier 2023
Take your YOLOv5 object detection projects to the next level.📈 With the new instance segmentation architectures, YOLOv5 now supports both segmentation and detection of objects.
To learn more, head over to https://t.co/XPIV4nt5ur #yolov5 #opencv #ai pic.twitter.com/mU1lHrt7m7Blog-tastic Tuesdays Take your YOLOv5 object detection projects to the next level. With the new instance segmentation architectures, YOLOv5 now supports both segmentation and detection of objects. To learn more, head over to https://
learnopencv.com/yolov5-instanc
e-segmentation/
… #yolov5 #opencv #ai -
TensorFlow Lite for Microcontrollers: Deploy TinyML Models
By
–
7. Deploying TinyML Learn to program in TensorFlow Lite for microcontrollers so that you can write the code, and deploy your model to your very own tiny microcontroller. https://
pll.harvard.edu/course/deployi
ng-tinyml?delta=0
…