Skip to main content

Home/ Advanced Concepts Team/ Group items tagged AI

Rss Feed Group items tagged

dharmeshtailor

FB pre-trained deep neural net on billion image user-hashtag dataset - 0 views

  •  
    Dataset automatically constructed from public images uploaded by users on FB/Instagram with hashtags used as labels! They refer to this as 'weakly supervised learning'. Then neural net fine-tuned for ImageNet and achieved record 85.4% accuracy.
Wiktor Piotrowski

Harnessing evolutionary creativity: evolving soft-bodied animats in simulated physical ... - 0 views

  •  
    Papers in the video description
jaihobah

Emergence of Locomotion Behaviours in Rich Environments - 1 views

shared by jaihobah on 11 Jul 17 - No Cached
jcunha liked it
  •  
    Some work by DeepMind on applying reinforcement learning to teach a computer to navigate complex environments. Come for the science - stay for the video: https://goo.gl/8rTx2F
LeopoldS

SpaceNet - 2 views

  •  
    nice, did any of you try it already
jaihobah

Artificial Neural Nets Grow Brainlike Navigation Cells - 0 views

  •  
    Faced with a navigational challenge, neural networks spontaneously evolved units resembling the grid cells that help living animals find their way.
Dario Izzo

Assured Autonomy - 5 views

  •  
    lol, while ESA is sending around memos and its managers are spending time talking about validation and verification of AI methods ... US / DARPA is already 5-6 years ahead. Hopefully the ACT can contribute to this with our DA based approach ....
  •  
    your call
dharmeshtailor

Facebook does Go - 3 views

  •  
    Took 2000 GPUs over 2 weeks to train :)
Marcus Maertens

Will California's New Bot Law Strengthen Democracy? | The New Yorker - 1 views

  •  
    California wants to force bots to reveal their identity in communication.
mkisantal

Robots Made Out of Branches Use Deep Learning to Walk - IEEE Spectrum - 1 views

  •  
    Random branches are collected, scanned to 3D, and connected with servos. Then a neural network is trained to control this "robot".
Marcus Maertens

Unsupervised word embeddings capture latent knowledge from materials science literature... - 1 views

  •  
    New results in NLP might allow to automate scientific discoveries by data mining of papers. Work considers 3.3M abstracts from material science, physics and chemistry and claims to discover new materials before they are published later on.
  •  
    ACT did that from diigo post digging in the retreat of 2014! Still without NLP.
  •  
    That's cool! Didn't know.
darioizzo2

Computer, enhance please! | Element AI - 3 views

  •  
    Blog from rarefin on our Proba-V Super Resolution competition
microno95

Introducing LCA: Loss Change Allocation for Neural Network Training | Uber Engineering ... - 2 views

  •  
    Fascinating insight into the question of how networks learn.
Dario Izzo

How the Space Pope is helping to find real exoplanets by playing Eve: Online | Ars Tech... - 0 views

  •  
    serious gaming came back!
jaihobah

New Theory Cracks Open the Black Box of Deep Learning - 0 views

  •  
    A new idea called the "information bottleneck" is helping to explain the puzzling success of today's artificial-intelligence algorithms - and might also explain how human brains learn.
jaihobah

A Brain Built From Atomic Switches Can Learn - 0 views

  •  
    A tiny self-organized mesh full of artificial synapses recalls its experiences and can solve simple problems. Its inventors hope it points the way to devices that match the brain's energy-efficient computing prowess.
pablo_gomez

RMA: Rapid Motor Adaptation for Legged Robots - 0 views

  •  
    71 Arguments why the ACT needs legged robots and a pretty impressive work.
microno95

Differences between deep neural networks and human perception | MIT News - 2 views

  •  
    The generated inputs are quite strange, I wonder where else something like this occurs.
LeopoldS

Fooling automated surveillance cameras: adversarial patches to attack person detection - 3 views

  •  
    beautiful! anybody knows of similar trials to fool EO data classifiers?
  •  
    I am not aware of any similar trials, yet this study remind me of generative adversarial talking heads https://arxiv.org/pdf/1803.07716.pdf https://www.youtube.com/watch?v=p1b5aiTrGzY
darioizzo2

Repulsive Curves | DeepAI - 0 views

  •  
    They pack a rabbit with an intestine!! Must be relevant to something ... and they plan to using the n-body code rakau developed by Francesco ....
darioizzo2

ESA Datalabs - 3 views

shared by darioizzo2 on 09 Mar 20 - No Cached
  •  
    This is the current page for ESA datalabs. Its an ongoing activity that we should monitor and join with Kelvins update if possible.
  •  
    I like the "computational narrative" term!
« First ‹ Previous 381 - 400 of 414 Next ›
Showing 20 items per page