DeepMind’s AI Plays Catch and So Much More!
Two Minute Papers examines the paper “Open-Ended Learning Leads to Generally Capable Agents.”
Read More
PonderNet: Learning to Ponder
-
Frank
- August 24, 2021
- AI
- Andrea Banino
- artificial intelligence
- Arxiv
- broader impact
- Charles Blundell
- Deep Learning
- deep learning classic algorithms
- deep learning recurrent computation
- deep network learning to stop
- DeepMind
- deepmind pondernet
- dynamic computation
- dynamic recurrent network
- explained
- halting probability
- Jan Balaguer
- Machine Learning
- Neural Networks
- Paper
- pondernet
- pondernet explained
- pondernet learning to ponder
Humans don’t spend the same amount of mental effort on all problems equally. Instead, we respond quickly to easy tasks, and we take our time to deliberate hard tasks. DeepMind’s PonderNet attempts to achieve the same by dynamically deciding how many computation steps to allocate to any single input sample. This is done via a […]
Read More
AlphaFold: The making of a scientific breakthrough
Here is the inside story of the DeepMind team of scientists and engineers who created AlphaFold, an AI system that is recognized as a solution to “protein folding”, a grand scientific challenge for more than 50 years.
Read More
DeepMind’s AI Watches YouTube and Learns To Play
Two Minute Papers looks at the paper “Playing hard exploration games by watching YouTube.”
Read More
NFNets: High-Performance Large-Scale Image Recognition Without Normalization
-
Frank
- February 15, 2021
- AI
- artificial intelligence
- Arxiv
- batch norm
- batch normalization
- batchnorm
- best imagenet model
- best neural network
- Deep Learning
- deep learning code
- deep learning tutorial
- Deep Mind
- DeepMind
- distributed training
- explained
- gradient clipping
- imagenet
- JAX
- layer normalization
- Machine Learning
- machine learning explained
- machine learning tutorial
- mean shift
- ml code
- Neural Networks
- nfnet
- nfnets
- nfnets code
- nfresnet
- normalizer-free
- Paper
- weight standardization
Yannic Kilcher explains NFNets in this video.
Read More
Did AI Solve the Protein Folding Problem?
Sabine Hossenfelder explains what the protein folding problem is, why it’s important, what AI means for biotech, and what the current situation is
Read More
Protein Folding Explained and Why DeepMind’s New Solution is a Big Deal
Earlier today, I shared Lex Fridman’s discussion on DeepMind’s recent advancement on protein folding. Join DeepMind Science Engineer Kathryn Tunyasuvunakool to explore the hidden world of proteins and why this discovery is a big deal. These tiny molecular machines underpin every biological process in every living thing and each one has a unique 3D shape […]
Read More
What DeepMind’s Protein Folding Breakthrough Means
Lex Fridman explains the world changing nature of DeepMind’s AlphaFold 2 breakthrough.
Read More
End-to-End Adversarial Text-to-Speech
-
Frank
- June 12, 2020
- adversarial
- AI
- aligner
- alignment
- artificial intelligence
- Arxiv
- convolutions
- Deep Learning
- Deep Mind
- DeepMind
- discriminator
- dynamic time warping
- explained
- GAN
- gaussian kernel
- Machine Learning
- mel
- Neural Networks
- Paper
- phonemes
- sound wave
- spectrogram
- Speech
- text-to-speech
- tokens
- tts
Text-to-speech engines are usually multi-stage pipelines that transform the signal into many intermediate representations and require supervision at each step. When trying to train TTS end-to-end, the alignment problem arises: Which text corresponds to which piece of sound? This paper uses an alignment module to tackle this problem and produces astonishingly good sound. Paper: https://arxiv.org/abs/2006.03575 […]
Read More
DeepMind Made A Superhuman AI For 57 Atari Games
Two Minute Papers takes a look at the paper “Agent57: Outperforming the Atari Human Benchmark” in the following video.
Read More