a bug-fix of QR-DQN network definition.

[ddlau]

[google-cla[bot]]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

:memo: Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here with @googlebot I signed it! and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

ℹ️ Googlers: Go here for more info.

[ddlau]

@googlebot I signed it!

[ddlau]

This is a Categorical-DQN legacy, which was just "transposed" in QR-DQN. In contrast to Categorical-DQN, QR-DQN set fixed probabilites and learnable locations (a.k.a diracs).

[psc-g]

Thanks for pointing this out! This is indeed semantically incorrect and is mostly a consequence of QR-DQN inheriting from Rainbow (which does expect probabilities). The probabilities field you rightly point out as not being probabilities is in fact not used at all in QR-DQN.

[ddlau]

rightly

Yeah, I noticed it too, after "uptraced" to its caller, and found the strange thing is not used. To be a little strict, isn't this sort of impreciseness and a misleading to new comings, more or less, I guess? After all, it is not proper, so I suggest the modification, at least some comment there.

[psc-g]

yup, we'll be adding some clarifications there. thanks for pointing this out!

On Tue, Nov 24, 2020 at 9:43 PM ddlau [email protected] wrote:

rightly

Yeah, I noticed it too, after "uptraced" to its caller, and found the strange thing is not used. To be a little strict, isn't this sort of impreciseness and a misleading to new comings, more or less, I guess? After all, it is not proper, so I suggest the modification, at least some comment there.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/google/dopamine/pull/157#issuecomment-733426860, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE3CCMN3PUXFERWCQHZ6UKLSRRVN7ANCNFSM4UAWCLAA .