Hutter prize: when will a compression method achieve 1 bit-per-character on a 100MB sample of Wikipedia?

Your submission is now in Draft mode. Once it's ready, please submit your draft for review by our team of Community Moderators. Thank you!

Question

The Hutter Prize is a 50,000€ Prize for Compressing Human Knowledge. The competition's stated mission is "to encourage development of intelligent compressors/programs as a path to AGI." Since it is argued that Wikipedia is a good indication of the "Human World Knowledge," the prize often benchmarks compression progress of algorithms using the enwik8 dataset, a representative 100MB extract from Wikipedia.

Since 2006, the Hutter Prize has galvanized not only data scientists but also many AI researchers who believe that image/text compression and AI are essentially two sides of the same coin. Compression algorithms are based on the premise of finding patterns in data and are predictive in nature. Furthermore, many machine learning researchers would agree that systems with better predictive models possess more "understanding" and intelligence in general.

The bits-per-character (the number of bits required per character) for compression of enwiki8 is the de-facto measurement unit for Hutter Prize compression progression. In 2016, the state of the art was set at 1.313 bits-per-character using Suprisal-Driven Zoneout, a regularization method for RNN.

We ask:

In what year will a language model generate sequences with less than 1.0 bits-per-character on the enwik8 dataset?

Resolution occurs when a method achieves less than 1.0 bits-per-character.

Make a Prediction

Prediction

Note: this question resolved before its original close time. All of your predictions came after the resolution, so you did not gain (or lose) any points for it.

Note: this question resolved before its original close time. You earned points up until the question resolution, but not afterwards.

This question is not yet open for predictions.

Current points depend on your prediction, the community's prediction, and the result. Your total earned points are averaged over the lifetime of the question, so predict early to get as many points as possible! See the FAQ.

Metaculus help: Predicting

Predictions are the heart of Metaculus. Predicting is how you contribute to the wisdom of the crowd, and how you earn points and build up your personal Metaculus track record.

The basics of predicting are very simple: move the slider to best match the likelihood of the outcome, and click predict. You can predict as often as you want, and you're encouraged to change your mind when new information becomes available.

The displayed score is split into current points and total points. Current points show how much your prediction is worth now, whereas total points show the combined worth of all of your predictions over the lifetime of the question. The scoring details are available on the FAQ.

Thanks for predicting!

Your prediction has been recorded anonymously.

Want to track your predictions, earn points, and hone your forecasting skills? Create an account today!

Track your predictions
Continue exploring the site