Search Results
| Search type | Search syntax |
|---|---|
| Tags | [tag] |
| Exact | "words here" |
| Author |
user:1234 user:me (yours) |
| Score |
score:3 (3+) score:0 (none) |
| Answers |
answers:3 (3+) answers:0 (none) isaccepted:yes hasaccepted:no inquestion:1234 |
| Views | views:250 |
| Code | code:"if (foo != bar)" |
| Sections |
title:apples body:"apples oranges" |
| URL | url:"*.example.com" |
| Saves | in:saves |
| Status |
closed:yes duplicate:no migrated:no wiki:no |
| Types |
is:question is:answer |
| Exclude |
-[tag] -apples |
| For more details on advanced search visit our help page | |
Results tagged with gradient
Search options not deleted
user 293973
Vector pointing in the direction where a function is growing fastest; its components are partial derivatives of this function. For questions about gradients in ecology, please use the [ecology] tag instead.
0
votes
Accepted
Need for reparameterization trick in RL (and others)?
The longer answer is (please feel free to correct me if I say something incorrect):
The gradient of the RL objective is of the form:
$$\nabla_\theta\mathbb{E}_{\tau\sim\pi(\theta)}[r(\tau)]=\nabla_\theta … tau)\nabla_\theta\pi(\theta)d\varepsilon=\mathbb{E}_{\tau\sim \varepsilon}[r(\tau)\nabla_\theta\pi(\theta)]$$
Intuitively, the reparameterization trick allows to separate the parameters with which the gradient …
- 145
0
votes
1
answer
418
views
Need for reparameterization trick in RL (and others)?
theta}=\dfrac{\partial \log \pi}{\partial \mu_\theta}=\dfrac{a-\mu_\theta}{\sigma}$$
1) Since an action $a$ has been sampled, can't that action $a$ be put back in $\nabla_\theta\log\pi$ an evaluate the gradient … For instance, just plugging in $a,s,\theta$ in the above equation would result in the gradient to update the mean. …
- 145