Lazy ABC
Duration: 39 mins 37 secs
Share this media item:
Embed this media item:
Embed this media item:
About this item
| Description: |
Prangle, D (University of Reading)
Thursday 24 April 2014, 14:20-14:55 |
|---|
| Created: | 2014-04-25 14:08 |
|---|---|
| Collection: | Advanced Monte Carlo Methods for Complex Inference Problems |
| Publisher: | Isaac Newton Institute |
| Copyright: | Prangle, D |
| Language: | eng (English) |
| Abstract: | In approximate Bayesian computation (ABC) algorithms, parameter proposals are accepted if corresponding simulated datasets are sufficiently close to the observations. Producing the large quantity of model simulations needed requires considerable computer time. However, it is often clear early on in a simulation that it is unlikely to produce a close match. This talk is on an ABC algorithm which saves time by abandoning such simulations early. A probabilistic stopping rule is used which leaves the target distribution unchanged from that of standard ABC. Applications of this idea beyond ABC are also discussed. |
|---|---|
Available Formats
| Format | Quality | Bitrate | Size | |||
|---|---|---|---|---|---|---|
| MPEG-4 Video | 640x360 | 1.93 Mbits/sec | 575.84 MB | View | Download | |
| WebM | 640x360 | 469.43 kbits/sec | 136.27 MB | View | Download | |
| iPod Video | 480x270 | 520.5 kbits/sec | 151.03 MB | View | Download | |
| MP3 | 44100 Hz | 249.73 kbits/sec | 72.56 MB | Listen | Download | |
| Auto * | (Allows browser to choose a format it supports) | |||||