Transkribus transcribing centuries-old letters and manuscripts
Long Short-Term Memory (LSTM)

Written by Quill, Read by Computer: An AI system for reading historical handwriting

The secrets of history are locked in troves of handwritten documents. Now a machine learning platform is making them amenable to digital search. Transkribus is transcribing centuries-old records en masse and making them available to scholars worldwide.
Data related to AI patents in the U.S.
Long Short-Term Memory (LSTM)

Charting the AI Patent Explosion: AI patents doubled between 2012 and 2019.

A new study used AI to track the explosive growth of AI innovation. Researchers from the U.S. Patent and Trademark Office deployed natural language processing to track AI’s increasing presence in four decades of patent data.
Alexa device and information about its new skill called natural turn-talking
Long Short-Term Memory (LSTM)

Alexa, Read My Lips: Amazon Alexa uses visual clues to determine who is talking.

Amazon’s digital assistant is using its eyes as well as its ears to figure out who’s talking. At its annual hardware showcase, Amazon introduced an Alexa skill that melds acoustic, linguistic, and visual cues to help the system keep track of individual speakers and topics of conversation.
Data related to a new reinforcement learning approach
Long Short-Term Memory (LSTM)

Eyes on the Prize: Vision-only reinforcement learning improves generalizability.

When the chips are down, humans can track critical details without being distracted by irrelevancies. New research helps reinforcement learning models similarly focus on the most important details.
Hamster running in a hamster ball
Long Short-Term Memory (LSTM)

Running Fast, Standing Still: Some state of the art machine learning progress is illusory.

Machine learning researchers report better and better results, but some of that progress may be illusory. Some models that appear to set a new state of the art haven’t been compared properly to their predecessors, Science News reports based on several published surveys.
Replica of the video game Pac-Man generated by a GAN
Long Short-Term Memory (LSTM)

Playing With GANs: GameGAN generated a fully functional Pac-Man.

Generative adversarial networks don’t just produce pretty pictures. They can build world models, too. A GAN generated a fully functional replica of the classic video game Pac-Man.
Single Headed Attention RNN (SHA-RNN)
Long Short-Term Memory (LSTM)

Language Modeling on One GPU: Single-headed attention competes with transformers.

The latest large, pretrained language models rely on trendy layers based on transformer networks. New research shows that these newfangled layers may not be necessary.
Comparison between TrXL and GTrXL
Long Short-Term Memory (LSTM)

Melding Transformers with RL

Large NLP models like BERT can answer questions about a document thanks to the transformer network, a sequence-processing architecture that retains information across much longer sequences than previous methods. But transformers have had little success in reinforcement learning — until now.
Process of labeling doctors' notes
Long Short-Term Memory (LSTM)

Cracking Open Doctors’ Notes

Weak supervision is the practice of assigning likely labels to unlabeled data using a variety of simple labeling functions. Then supervised methods can be used on top of the now-labeled data.
Schematic of the architecture used in experiments related to systematic reasoning in deep reinforcement learning
Long Short-Term Memory (LSTM)

How Neural Networks Generalize

Humans understand the world by abstraction: If you grasp the concept of grabbing a stick, then you’ll also comprehend grabbing a ball. New work explores deep learning agents’ ability to do the same thing — an important aspect of their ability to generalize.

Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox