Loading
0 Follower
0 Following
BrandonHoughton
Brandon Houghton

Organization

MineRL Labs - Carnegie Mellon University

Location

US

Badges

1
0
0

Activity

Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Mon
Wed
Fri

Ratings Progression

Loading...

Challenge Categories

Loading...

Challenges Entered

Latest submissions

No submissions made in this challenge.

Latest submissions

No submissions made in this challenge.

Sample Efficient Reinforcement Learning in Minecraft

Latest submissions

No submissions made in this challenge.

Measure sample efficiency and generalization in reinforcement learning using procedurally generated environments

Latest submissions

No submissions made in this challenge.

Sample-efficient reinforcement learning in Minecraft

Latest submissions

No submissions made in this challenge.

Sample-efficient reinforcement learning in Minecraft

Latest submissions

No submissions made in this challenge.

Multi Agent Reinforcement Learning on Trains.

Latest submissions

No submissions made in this challenge.
Participant Rating
Participant Rating
BrandonHoughton has not joined any teams yet...

NeurIPS 2020: MineRL Competition

Am I allowed to use the openai gym environment without joining the competition?

Over 1 year ago

The MineRL package is open-source and we certainly welcome anyone to experiment with it!

Division of datasets based on reward

Almost 2 years ago

Hi @michael-tanaka,

Thanks for your question! This year we decided that you CAN use reward when learning the action distribution from human demonstrations. E.g. it is permitted to learn the joint distribution between reward and human actions and condition on this distribution when sampling.

What you are describing, however, sounds like a hard-coded meta-controller, as the policy is dictated by hand-encoding the reward thresholds.

One option to mitigate this would simply be to learn a meta-controller that only observes reward, and decides against a fixed number of policies. You could then weight demonstrations by their reward to have a uniform sampling distribution.

The difference between ObtainDiamondVectorObf and ObtainDiamondDenseVectorObf

Almost 2 years ago

That’s an error with the documentation - I thought we had fixed that but we must have missed a section, sorry!

NeurIPS 2019 : MineRL Competition

Cannot download the dataset

Over 2 years ago

Should be working now!

About the rule on pre-trained model

Almost 3 years ago

Miffyli is correct here - even pre-training using a small number of learned weights is not allowed.

We will be investigating code as well to validate submissions, the large file restriction simply provides an easy way to enforce the pre-training rule generally

When will the results of round 1 be announced?

Almost 3 years ago

Teams should be notified! Congratulations to the top teams!

Can't train in MineRLObtainIronPickaxeDense-v0 since 0.2.7

Almost 3 years ago

Fixed! Install minerl 0.2.8

Can't train in MineRLObtainIronPickaxeDense-v0 since 0.2.7

Almost 3 years ago

Great catch - until we can update the PyPI repo, using the MineRLObtainDiamondDense-v0 environment should be a close replacement especially if you limit the number of steps!

Can't train in MineRLObtainIronPickaxeDense-v0 since 0.2.7

Almost 3 years ago

Sorry I will take a look now - I thought this was covered by our unit tests!

Announcement - Round 1 Scores

Almost 3 years ago

To clarify - we now have moved to Round 1.5! The scores of Round 1 will be for archival purposes only.

Announcement - Round 1 Scores

Almost 3 years ago

Announcement - Round 1 Scores

We have reviewed multiple submissions that obtain rewards that should not be achievable in the MineRLObtainDiamond-v0 environment.

As this is due to an easily exploitable reward loop present in outdated minerl versions (prior to minerl 0.2.5,) we have decided to add 5 additional submissions to each team. The new maximum number of submissions is now 25.

Please verify submissions locally to ensure your current scoreboard results. Top submissions submitted using out-dated minerl versions (prior to minerl 0.2.5) will be re-run to verify their performance.

Additionally, participants should retrain their models to account for the reward loop removal.

On AIcrowd, the minerl version can be checked by looking for the minerl==<version> line in the requirements.txt file.

Locally, the python package can be updated with python -m pip install --upgrade minerl command.

Internal Reward Dependent on Expert Data and State

Almost 3 years ago

As long as the internal reward is learned from the data, this is allowed. This is not allowed if it is directly a function of the state and external data.

Did the new dataset release already?

Almost 3 years ago

Unfortunately we are unable to release additional data at this time.
We will make an announcement if more data will be available for round 2.

Repeated reward for logs and furnace

Almost 3 years ago

Just to follow up here - this was indeed an issue and the fix is being bundled in minerl 0.2.5!

Equip item failed

Almost 3 years ago

This was an issue with the obtainDiamond.xml - we have resolved it in the most recent release being deployed today or tomorrow!

Using open-sourced networks

Almost 3 years ago

Unfortunately, ImageNet pre-training is not allowed this year!

Re-training models is a key part of round 2 and if pre-trained weights are used there is no way to tell how those pre-trained weights were generated. Additionally, if pre-training happened during evaluation, it would be possible for competitors to upload large amounts of data which could be used to load other pre-trained weights.

In future iterations of the competition, if pre-training on ImageNet is a common ask, we could consider including certain datasets in the provided docker container; however, note that the texture pack of Minecraft will change in round 2 so techniques that work well transferring from the natural images to Minecraft may not work well in round 2!

How to use furnace to "nearbySmelt" coal?

Almost 3 years ago

They should have the same item-ID so this should not be an issue but I will verify this when checking it out!

Agent Behavior for conflicting actions?

Almost 3 years ago

The behavior is defined as occurs in vanilla Minecraft (where possible). For movement, nothing happens when asking for conflicting actions. For place and attack, both actions will be processed as the place handler is through Malmo and the attack action is handled by default Mincraft.

How to use furnace to "nearbySmelt" coal?

Almost 3 years ago

Thanks for this, I will take a look. Could be a weird interaction between the Minecraft give commands and the Malmo agent, I will explore building a world with the needed resources and see if this is still the case

NeurIPS 2019 - Robot open-Ended Autonomous Lear...

Deadline for round 1

Almost 3 years ago

The deadline has been extended previously as announced:
Sep 22, 2019 Oct 25, 2019 (UTC 12:00):
Check your system time, perhaps your date is set improperly. If not please follow up with @mohanty!

Organizer for MineRL Competition on Sample Efficient Reinforcement Learning