Sampling is also ubiquitous in science, engineering, data analysis, etc. etc.
Are there other examples that you have encountered in your studies and projects?
bronyayang
In machine learning, sampling from a distribution is used for creating a dataset and helping downstream tasks.
CarneAsadaFry
I do research in nuclear physics, where we attempt to characterize the properties of scintillators (materials that emit light when exposed to ionizing radiation). We shoot neutrons at the scintillator, and then try to quantify the light emitted by the scintillator using devices called photomultiplier tubes (PMTs) that collect the emitted photons and produce electrical signals. However, we can really only sample the electrical signals the PMTs produce, and we must account for this when analyzing the data.
rubywerman
When designing filters in electrical engineering, it's important to consider the sampling rate. One example of this is creating a band-pass filter for human-voice, like we did in EECS16B
LinyueSong
Obviously, sampling is also important in statistics. When we want to investigate some characteristics of the population, we often have to take a sample because the population size is too large. There are typically for types of sampling methods: simple random sampling, systematic sampling, stratified sampling, and clustered sampling. I'm sure a lot of us have learned these in some stats classes. Wonder if any of these methods will be used in this class
BohanYu
AI also uses lots of sampling techniques like Bayes net sampling (e.g. prior sampling, rejection sampling, likelihood weighting, gibbs sampling) and particle filtering, which helps solving robot localization / SLAM problems.
adityaramkumar
Sampling is also used in some scheduling algorithms for computer systems. In particular, we may think about how we can schedule a set of n random threads, where each thread is scheduled ki/n time, where each ki can be distinct. This is called lottery scheduling, and entails sampling tickets.
Sampling is also ubiquitous in science, engineering, data analysis, etc. etc.
Are there other examples that you have encountered in your studies and projects?
In machine learning, sampling from a distribution is used for creating a dataset and helping downstream tasks.
I do research in nuclear physics, where we attempt to characterize the properties of scintillators (materials that emit light when exposed to ionizing radiation). We shoot neutrons at the scintillator, and then try to quantify the light emitted by the scintillator using devices called photomultiplier tubes (PMTs) that collect the emitted photons and produce electrical signals. However, we can really only sample the electrical signals the PMTs produce, and we must account for this when analyzing the data.
When designing filters in electrical engineering, it's important to consider the sampling rate. One example of this is creating a band-pass filter for human-voice, like we did in EECS16B
Obviously, sampling is also important in statistics. When we want to investigate some characteristics of the population, we often have to take a sample because the population size is too large. There are typically for types of sampling methods: simple random sampling, systematic sampling, stratified sampling, and clustered sampling. I'm sure a lot of us have learned these in some stats classes. Wonder if any of these methods will be used in this class
AI also uses lots of sampling techniques like Bayes net sampling (e.g. prior sampling, rejection sampling, likelihood weighting, gibbs sampling) and particle filtering, which helps solving robot localization / SLAM problems.
Sampling is also used in some scheduling algorithms for computer systems. In particular, we may think about how we can schedule a set of n random threads, where each thread is scheduled ki/n time, where each ki can be distinct. This is called lottery scheduling, and entails sampling tickets.