Neural networks have learned to play video games like Dota 2 via reinforcement learning by playing for the equivalent of thousands of years (compressed into far less time). In new work, an automated player learned not by playing for millennia but by watching a few days’ worth of recorded gameplay.

Getting high accuracy out of a classifier trained on a small number of examples is tricky. You might train the model on several large-scale datasets prior to few-shot training, but what if the few-shot dataset includes novel classes? A new method performs well even in that case.

Using a neural network to generate realistic videos takes a lot of computation. New work performs the task efficiently enough to run on a beefy personal computer.

Systems designed to turn handwriting into text typically work best on pages with a consistent layout, such as a single column unbroken by drawings, diagrams, or extraneous symbols. A new system removes that requirement.

Researchers have used neural networks to generate novel views of a 3D scene based on existing pictures plus the positions and angles of the cameras that took them. In practice, though, you may not know the precise camera

Researchers have proposed task-agnostic architectures for image classification tasks and language tasks. New work proposes a single architecture for vision-language tasks.

Transformer models trained on sequences of amino acids that form proteins have had success classifying and generating viable sequences. New research shows that they also capture information about protein structure.

Researchers typically downsize images for vision networks to accommodate limited memory and accelerate processing. A new method not only compresses images but yields better classification.

Ideally, real-time 3D applications such as virtual and augmented reality transition smoothly between different viewpoints of a scene — but generating a fresh perspective can take time. New research speeds the process.

In the open-ended video game Minecraft, players extract blocks of virtual materials from a 3D environment to assemble objects of their own design, from trees to cathedrals. Researchers trained neural networks to generate these structures.

IBM’s Watson question-answering system stunned the world in 2011 when it bested human champions of the TV trivia game show Jeopardy! Although the Watson brand has fallen on hard times, the company’s language-processing prowess continues to develop.

Researchers at the University of Bristol led by Michael Wray propose a new benchmark, Semantic Similarity Video Retrieval (SVR), that evaluates video retrieval systems by their ability to rank many similar videos. They also built a system that performed well on it.

In some animated games, different characters can perform the same actions — say, walking, jumping, or casting spells. A new system learned from unlabeled data to transfer such motions from one character to another.

A recent generative adversarial network (GAN) produced more coherent images using modified transformers that replaced fully connected layers with convolutional layers. A new GAN achieved a similar end using transformers in their original form.

It’s well established that pretraining a model on a large dataset improves performance on fine-tuned tasks. In sufficient quantity and paired with a big model, even data scraped from the internet at random can contribute to the performance boost.