The Next Input updates 1,127 published updates

The Next Input updates

Browse every published The Next Input update in a calm card overview with images, dates, and direct access to each article.



The Next Input update

The Next Input
29 Sep 2017

Nonlinear computation in deep linear networks

Open article →

The Next Input update

The Next Input
14 Sep 2017

Learning to model other minds

Read paper(opens in a new window)(opens in a new window)

Open article →

The Next Input update

The Next Input
13 Sep 2017

Learning with opponent-learning awareness

Read paper(opens in a new window)

Open article →

The Next Input update

The Next Input
18 Aug 2017

OpenAI Baselines: ACKTR & A2C

Read code(opens in a new window)Read paper(opens in a new window)

Open article →

The Next Input update

The Next Input
16 Aug 2017

More on Dota 2

Our Dota 2 result shows that self-play can catapult the performance of machine learning systems from far below human level to superhuman, given sufficient compute. In the span of a month, our system went from barely matching a high-ranked player to beating the top pros and has continued to improve since then. Supervised deep learning systems can only be as good as their training datasets, but in self-play systems, the available data improves automatically as the agent gets better.

Open article →

The Next Input update

The Next Input
11 Aug 2017

Dota 2

Rewatch live event

Open article →

The Next Input update

The Next Input
3 Aug 2017

Gathering human feedback

View code(opens in a new window)

Open article →

The Next Input update

The Next Input
27 Jul 2017

Better exploration with parameter noise

Read code(opens in a new window)Read paper(opens in a new window)

Open article →

The Next Input update

The Next Input
20 Jul 2017

Proximal Policy Optimization

View code(opens in a new window)Read paper(opens in a new window)

Open article →

The Next Input update

The Next Input
17 Jul 2017

Robust adversarial inputs

We’ve created images that reliably fool neural network classifiers when viewed from varied scales and perspectives. This challenges a claim from last week that self-driving cars would be hard to trick maliciously since they capture images from multiple scales, angles, perspectives, and the like.

Open article →

The Next Input update

The Next Input
5 Jul 2017

Hindsight Experience Replay

Open article →

The Next Input update

The Next Input
1 Jul 2017

Teacher–student curriculum learning

Read paper(opens in a new window)

Open article →

Showing 1021 to 1032 of 1,127 updates.

Gemini komt eraan