(I work at OpenAI) We're planning to cover the cost for fine-tuning replacement models. We're still working through the exact mechanics that will work best for customers, and will be reaching out to customers to get feedback on different approaches in the next few weeks.
(author here) It depends a bit on how you define "retain". Most GPT-3 applications use custom "prompts" to train it for their specific use case. So in that way, the prompt is retained with every request.
You can also fine-tune GPT-3 to retain the ability to reason through problems. For example, check out this work on reasoning for grade school math: https://openai.com/blog/grade-school-math/
We use W&B a lot for our research projects, and it's been super useful and has pretty much replaced TensorBoard (TB) for us. The killer features on top of TB is the ability to track your experiments (without keeping track of your log files on your computer -- i'm quite organized but that's a losing battle), and the ability to share training runs within a team. We do a lot of distributed training as well, and the system metrics they provide automatically are great for ensuring you're making the most of your GPU/CPU/memory.
The vision network is trained before-hand on lots of different configurations in simulation and then used to infer the block locations in the image from the camera. So it’s not learning continously. The imitation network takes the block locations predicted by the vision network, together with the demonstration trajectory in VR, and imitates the task shown in the demonstration. So, it learns to look through the demonstration to decide what action to take next given the current state (i.e. location of blocks and gripper). To keep the setup simple, we only trained the imitation network on stacking tasks (so no unstacking, throwing, etc). In future work, we want to make the setup and tasks much more general.
We actually use both. The data used to train the neural networks is all generated through a scripted policy simulation. We then use a single human demonstration in VR to show the robot how to carry out a particular task. VR lets a human do a demonstration quite naturally, and would also be an effective way to collect training data for more complex tasks, where we’re not be able to create a scripted policy.