feat: Add 2020 roadmap

[gaocegege]

Signed-off-by: Ce Gao gaoce@caicloud.io

What this PR does / why we need it:

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Fixes #

Special notes for your reviewer:

1. Please confirm that if this PR changes any image versions, then that's the sole change this PR makes.

Release note:

[kubeflow-bot]

This change is 

[gaocegege]

/assign @johnugeorge @andreyvelich

[johnugeorge]

/lgtm

[andreyvelich]

@gaocegege Thank you for the PR. I added few things.

[andreyvelich]

Maybe also add: Change design of NAS Model Constructor?

As you know, we have model constructor inside Training Container. I think it is not very useful for the user.

If we can construct model somewhere else (Suggestion or Init Container) and send, for example, json representation of the model directly to the training container, user doesn't need to include this part to the training container.

I will create an issue about it soon.

[gaocegege]

LGTM. Looking forward to your issue.

[andreyvelich]

Also, I found that we have something in TODO for the current NAS RL suggestion: https://github.com/kubeflow/katib/tree/master/pkg/suggestion/v1alpha3/NAS_Reinforcement_Learning#to-do.

What do you think, should we add something to the ROADMAP?
For example, step 1: "Add 'micro' mode, which means searching for a neural cell instead of the whole neural network. "

/cc @johnugeorge @gaocegege

[k8s-ci-robot]

@andreyvelich: GitHub didn't allow me to request PR reviews from the following users: gaocegege.

Note that only kubeflow members and repo collaborators can review this PR, and authors cannot review their own PRs.

In response to this:

Also, I found that we have something in TODO for the current NAS RL suggestion: https://github.com/kubeflow/katib/tree/master/pkg/suggestion/v1alpha3/NAS_Reinforcement_Learning#to-do.

What do you think, should we add something to the ROADMAP?
For example, step 1: "Add 'micro' mode, which means searching for a neural cell instead of the whole neural network. "

/cc @johnugeorge @gaocegege

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[gaocegege]

@andreyvelich Do we have the bandwidth for the TODO in ENAS? I think we can add it but in low priority. WDYT

[gaocegege]

Comments are addressed, PTAL

/assign @andreyvelich

[johnugeorge]

/lgtm

[johnugeorge]

Fixes: #1104

[johnugeorge]

/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: johnugeorge

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [johnugeorge]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[andreyvelich]

@gaocegege Thank you for the changes!
Regarding to ENAS, yes I have few thoughts that can be improved.
Should I add them to the ROADMAP or issue is enough?