9 votes
Accepted

How much of Deep Mind's work is actually reproducible?

On the suggestion of the O.P. rcpinto I converted a comment about seeing "around a half-dozen papers that follow up on Graves et al.'s work which have produced results of the caliber" and will ...
user avatar
8 votes

Is AlphaFold just making a good estimate of the protein structure?

AlphaFold (version 1 and 2) predicts (so estimates) the 3D shape of the protein from the sequence of amino acids. AlphaFold's performance is measured with the global distance test (GDT), which is a ...
user avatar
  • 33k
5 votes
Accepted

Why did AlphaGo lose its Go game?

We know what Lee's strategy was during the game, and it seems like the sort of thing that should work. Here's an article explaining it. Short version: yes, we know what went wrong, but probably not ...
user avatar
4 votes

How much of Deep Mind's work is actually reproducible?

I tend to think this question is border-line and may get close. A few comments for now, though. wrongx There are (at least) two issues with reproducing the work of a company like DeepMind: ...
user avatar
  • 1,480
4 votes

What's the difference between Starcraft and Dota from an AI perspective?

In SC2, players have more control over every minuet mechanic (constructing buildings, resource mining and management, controlling minions...) in the game, thus putting more tactical responsibility on ...
user avatar
4 votes

What's the difference between Starcraft and Dota from an AI perspective?

Those AI-learning programs may have very similar scheme. We are changing only inputs and possible actions (like "use skill" or "move here"). Starcraft AI must do a lot of actions and control many ...
user avatar
3 votes
Accepted

Is it common in RL research with Atari/ALE to automatically press FIRE to start games?

Based on this comment in the Issue I created about this question on github, it looks like there is confirmation that at least DeepMind does not use this kind of functionality in their Atari ...
user avatar
  • 9,316
3 votes
Accepted

Was DeepMind's DQN learning simultaneously all the Atari games?

Switching required a re-learn. Also, note that: We use the same network architecture, learning algorithm and hyperparameters settings across all seven games, showing that our approach is robust ...
user avatar
3 votes

Alpha Zero queen promotion

It means that there is no explicit coding of action choices to promote to queen, it is the default assumption if the underpromotion actions are not taken. The Alpha Zero chess implementation can ...
user avatar
  • 23.1k
3 votes

How does DeepMind perform reinforcement learning on a TPU?

In their blog post, they link to (among many other papers) their IMPALA paper. Now, the blog post only links to that paper with text implying that they're using the "off-policy actor-critic ...
user avatar
  • 9,316
3 votes
Accepted

Each training run for DDQN agent takes 2 days, and still ends up with -13 avg score, but OpenAi baseline DQN needs only an hour to converge to +18?

Although what @Jaden said may be true by itself, it does not really serve to answer my question as I have seen after conducting numerous experiments, and finally reaching close to Dueling Network ...
user avatar
  • 203
3 votes

Each training run for DDQN agent takes 2 days, and still ends up with -13 avg score, but OpenAi baseline DQN needs only an hour to converge to +18?

Dueling architectures create bigger differences in the values of actions in the state space. This is because the state-value V(s) function is estimated separately from the state-action value Q(s, a). ...
user avatar
2 votes

3D environment for RL research in Academia

For example DeepMind [Lab] is not a standard or Open Source friendly I'm not sure where you got that info from... as far as I'm aware, DeepMind Lab is definitely used in various publications (maybe ...
user avatar
  • 9,316
2 votes

Was DeepMind's DQN learning simultaneously all the Atari games?

Switching requires relearning, the network did not have a single set of weights that allowed it to play all games well. This is due to the catastrophic forgetting problem. However, recent work has ...
user avatar
  • 21
1 vote
Accepted

How does the AlphaGo Zero policy decide what move to execute?

The formula in question uses a function N(state, action) that defines a visit count of a state-action pair (introduced on page 3). To describe how it is used, lets first describe the steps of AlphaGo ...
user avatar
1 vote
Accepted

Which simulation platform is used by DeepMind (and others) to handle inverse kinematics musculoskeletal?

DeepMind used MuJoCo (see also the related paper MuJoCo: A physics engine for model-based control) for the simulations, as they stated in section 3.1 of their paper Emergence of Locomotion Behaviours ...
user avatar
  • 33k
1 vote
Accepted

What is regression layer in a spatial transformer?

Basically, it means that the "localization network" should output a set of real valued parameters (typically 6 numbers). The word "regression" doesn't bear any specific meaning. Any network that ...
user avatar
  • 120

Only top scored, non community-wiki answers of a minimum length are eligible