[Feature Request] Implement TD3Trainer #3558

Motivation

Many of the most popular RL algorithms have already been implemented in the trainers framework (thank you!!). It would be great to also add TD3!

Solution

PR here: #3557

Alternatives

N/A

Additional context

Checklist

• I have checked that there is no similar issue in the repo (required)