This is the Relational DBN for the interruption model This is the domain of A. This is the domain of S. This is the domain of I....

There are now four models that I have tested with this data. The first model is a simple HMM that is event driven: activities are only allowed to switch when a tag is seen. This is called the "Untimed"...

Here is some data that shows how well four different distributions match the true timing distribution in the data. To get these numbers, I took all the activities that were in the ADL dataset from Matthai's house and I...

So what are the conclusions. Well, exponential timing distributions don't improve accuracy. That seems like a reasonable conclusion since the distributions are not exponential. Would a different timing distribution help things?...

So here is what I think is reasonable to conclude from this data. Nothing earth shattering, but now there is data to support it: Activity durations are not well characterized by exponential distributions. One might model them differently and possibly...

Here are three graphs comparing the probability of seeing an object given an activity. This first one compares engaging in a leisure activity with making a snack. In this case there is little in common between the two profiles. As...

Here are some results for using HMM's to recognize ADLs. First about the data set: This was the dataset that Matthai first recorded at his house. It consists of 17 non-researchers performing 12 activities one from each of 12 classes....

Matthai has a nice list of tasks in activity inference that are increasingly difficult: Logging: What happened today? Rating: How well was it done? Trending: Is the performance changing over time? Evaluation: What did you do wrong/right? Guidance: What are...

6 items touched 4 items touched 5 items touched 3 items touched 3 items touched 2 items touched 2 items touched 3 items touched 2 items touched 1 items touched 4 items touched 3 items touched 2 items touched...

Here are my stats from the data set after Marie was done with recoding the video: There are 17 traces from 8 doctors. One doctor was a novice (who produced one trace). There were 55 tags seen that we could...

This is an application of activity inference that we have been working on at Intel. The basic problem is to assign a grade to someone's performance of an activity based on observing an RFID sensor stream. In particular we...

Here is the data collected from the anesthesiologists run. It consists of 17 traces with RFID tag reads from two gloves. The alien reader tags have been removed. It is missing the English interpretation of 19 tags which aren't...

In a previous article I discussed some challenges associated with trying to grade to sequences of activities. One of which is the true, or gold trace, and the other is the inferred, or black, trace. In that discussion I...

From Decision-Theoretic Planning: Structural Assumptions and Computational Leverage: A classical STRIPS operator is described by a precondition and a set of effects. The former identifies the set of states in which the action can be executed, and the latter...

Here is a small interesting problem. A user in a smart home performs a sequence of activities and the smart home records a sensor stream that was generated by those activities. The sensors might be motion sensors or RFID...

After a meeting yesterday with Henry, I think that we convinced ourselves that we have the details of the activity inference model worked out. For the benefit of the hordes of people who would like to know the structure...

The following is my summary of the paper reviews that we received for the RFID Activity Inference Paper submission to UBICOMP (that was, unfortunately, rejected). The biggest issue that is recurrently coming up is why we need so much...

This paper describes a new way of modeling the changes in a state space over time called Continuous Time Bayesian Networks (CTBNs). It uses a novel graphical model framework which competes with a Dynamic Bayesian Network and is specifically...

I just finished reading A Closed-Form Solution for Mapping General Distributions to Minimal PH Distributions. This was a paper that came from CMU and shows how one can approximate any member of a family of functions. The family of...