You are viewing a plain text version of this content. The canonical link for it is here.
Posted to discuss-archive@tvm.apache.org by Pzq via Apache TVM Discuss <no...@discuss.tvm.ai> on 2021/08/20 10:34:22 UTC
[Apache TVM Discuss] [Questions] [AutoTVM] Question about Bayesian
Optimization
According to the paper "Learning to Optimize Tensor Programs", it seems that Bayesian Optimization is not a good choice as a tuner because of the reasons shown below.
1. Uncertainty estimation was not as important in autotuning problem, possibly because the models were trained with more training samples than traditional hyper-parameter optimization problems.
2. Configuration space s is not invarient which makes Bayesian Optimization not working on transfer learning.
Am I correct?
I took screenshots of the paragraphs in the paper.
So Bayesian Optimization do not work well on auto tuning tasks, why It was mentioned in the last section of the paper?
---
[Visit Topic](https://discuss.tvm.apache.org/t/autotvm-question-about-bayesian-optimization/10852/1) to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/437c5954f3475961e95303e89f09a9d513621c39588ffcb71b0f59a7fa000276).