Comments (5)
Ok, everything is clear now. Thank you very much for your time.
Have a good day,
Jurijs
from torchsde.
Hi Jurijs,
Thanks for the interest. The other term you're mentioning (the term is actually an expected integral over 1/2*((f-h)/g)^2
, as this term on its own doesn't have much meaning) is in fact a Monte Carlo estimator of a KL divergence on path space. There are two SDEs involved here: One we call the prior, and the other we call the approx. posterior.
Each of the two SDEs induce their own solution, each of which in turn is a stochastic process and defines a distribution over the space of functions (e.g. C([0, 1], R^d)
). With these two distributions, we can now define a KL divergence, and estimate it using Monte Carlo. This procedure results in that term.
Obviously, what I'm claiming here simplifies some technical aspects, and I'd recommend our paper section 9.6 (and the relevant section in the main text) for a detailed derivation. The neural SDE paper by Tzen & Raginsky also has relevant information. The chapter on Girsanov's theorem/likelihood-ratio of Ito process in the Applied SDE book also has relevent information, though the derivation is largely heuristic and technically a bit off on the math details. The derivation there is based on viewing SDEs as the limit of incrementing scaled Gaussians, so it's by the far the most approachable.
from torchsde.
Closing this now. Feel free to reopen if there are additional questions.
from torchsde.
Hi Xuechen,
Thanks for the response, it was very helpful. I have 2 additional questions regarding your paper/code. It would be wonderful, if you reply.
- In section 9.6.1 the derivation of ELBO ends up with expected value w.r.t. P-law, which is law of prior. However, in the section 5 and in the code, the expected value is computed w.r.t. approximate posterior distribution Q.
- Despite the question 1) I understand the derivation of ELBO and appearance of loop_path in the code. However, it is not clear for me where is the standard KL term between prior and posterior in the ELBO, notated as loqp0 in the code.
Jurijs
from torchsde.
In section 9.6.1 the derivation of ELBO ends up with expected value w.r.t. P-law, which is law of prior. However, in the section 5 and in the code, the expected value is computed w.r.t. approximate posterior distribution Q.
P and Q in our paper are prob. measures on the underlying prob. space. With this measure, W_t is a standard Wiener process (whereas hat{W_t} is not). It is not the law/distribution induced by the stochastic process (we call this the P-law of the stoch. process, if P is the prob. measure; similarly, if Q is the prob. measure, then the law induced by the stoch. process is called its Q-law). Clearly, if you change the underlying law on the prob. space to something other than P, W_t may not be a standard Wiener process anymore, and hence the distribution on path space induced by the SDE would also change.
So under P, W_t is a standard Wiener process, therefore we could simulate the SDE with h_phi and sigma using sdeint.
Despite the question 1) I understand the derivation of ELBO and appearance of loop_path in the code. However, it is not clear for me where is the standard KL term between prior and posterior in the ELBO, notated as loqp0 in the code.
KL at the time=0 was omitted in the paper to save space.
from torchsde.
Related Issues (20)
- Using "*" in version speficiers is deprecated and breaks some installs HOT 3
- Defining noise type for vector-valued homogeneous SDE HOT 1
- vector-valued SDE cumbersome workflow HOT 1
- Zero drift and zero diffusion matrices lead to non-zero changes of variable
- Irregular data and sampling posterior in latent_sde_lorenz.py
- Low CPU and GPU usage in training sde_gan, Seeking Help to Improve Performance. HOT 2
- Different `t` for data in a minibatch HOT 5
- torchsde pypi package is misformated HOT 26
- Deprecation torchsde version HOT 5
- Something went wrong Expecting value: line 1 column 1 (char 0) HOT 1
- Incorporating real stock time series data
- SDE-LSTM structure for time series forecasting
- Performance Improvement Inquiry: Experiencing Slow Execution with TorchSDE HOT 2
- Latent SDE failed to generate longer samples
- issue with my text to image ai Device type privateuseone is not supported for torch.Generator() api. HOT 1
- 我的 Mac上 只有torchsde-0.2.6.dist-info,怎样才能安装insightface
- 如果 torchsde 当前没有解决这个问题的新版本,你可以联系该项目的维护者或作者
- sdeint_adjoint for f_and_g_prod type SDE, how to set parameter?
- failure to install pip install qiskit-aer to run qsvm on ibm simulator
- learning the generative model of periodical process HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from torchsde.