F.V. Jensen, U. Kjrulff, K.G. Olesen, and J. Pedersen. An expert system for control of waste water treatment---a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, 1989. In Danish. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Decayed MCMC Filtering - Marthi, Pasula, Russell, Peres (2002) (2 citations) (Correct)
....hand, in Figure 4, the situation is reversed, and the fast exponential outperforms the slow one, because in this case it is a better match for the forgetting rate of the DBN. The quadratic decay is more robust, performing well for both HMMs. We next consider a larger DBN the WATER network [Jensen et al. 1989] , used for monitoring a water purification plant. Figure 5 shows error as a function of history length, using 1000 samples. Undecayed MCMC shows the expected increase in error, as the samples are forced to cover more ground. Among the other algorithms, fixedwindow MCMC does slightly worse than ....
F. Jensen, U. Kjaerulff, K. Olesen, and J. Pedersen. An expert system for control of waste water treatment - a pilot project. Technical report, Judex Datasystemer, Aalborg, Denmark, 1989.
Sampling in Factored Dynamic Systems - Koller, Lerner (2000) (6 citations) (Correct)
....both discrete and continuous variables. Hidden Markov models are a very simple special case of DBNs, as are linear Gaussian systems (Kalman 1960, Bar Shalom and Fortmann 1988) DBNs have been used for a variety of applications, including freeway surveillance (Huang et al. 1994) complex factories (Jensen, Kjaerulff, Olesen and Pedersen 1989), robotics (Nicholson and Brady 1994) medical monitoring (Dagum and Galper 1995) speech recognition (Zweig and Russell 1998) and more. Of course, modeling complex systems is only the first step; we also need to use these models for inference. A common goal in dynamic systems is tracking (also ....
Jensen, F., Kjaerulff, U., Olesen, K. and Pedersen, J. (1989). An expert system for control of waste water treatment--- a pilot project, Technical report, Judex Datasystemer A/S, Aalborg, Denmark.
Tractable Inference for Complex Stochastic Processes - Boyen, Koller (1998) (96 citations) (Correct)
....ideas in the context of a monitoring task, showing that orders of magnitude faster inference can be achieved with only a small degradation in accuracy. page 1 of 1 Introduction The ability to model and reason about stochastic processes is fundamental to many applications [ Forbes et al. 1995; Jensen et al. 1989; Dagum et al. 1992; Provan, 1992 ] For example, we may be observing a freeway traffic scene via a video camera mounted on a bridge, with the goal of understanding the current traffic situation and predicting its future evolution [ Forbes et al. 1995 ] or monitoring a patient s symptoms to ....
....them to be independent. Specifically, if we have two (sets of) variables that are highly correlated, splitting them into two separate subprocesses is probably not a good idea. 5. 3 Experimental results We have validated this algorithm in the context of two real life DBNs: the water network [ Jensen et al. 1989 ] used for monitoring the biological processes of a water purification plant; and the bat network [ Forbes et al. 1995 ] used for monitoring freeway traffic (see Figure 4) We have added a few evidence nodes to water, which did not have any; these duplicate a few of the state variables with ....
F.V. Jensen, U. Kjaerulff, K.G. Olesen, and J. Pedersen. An expert system for control of waste water treatment--- a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, Denmark, 1989. In Danish.
Using Learning for Approximation in Stochastic Processes - Koller, Fratkina (1998) (33 citations) (Correct)
.... over states at time t 1 given the state at time t is represented in a network fragment such as the one in Figure 1(a) appropriately annotated with probabilities) DBNs have been used for a variety of applications, including freeway surveillance [FHKR95] monitoring complex factories [JKOP89], and more. Exact inference algorithms for BNs have analogues for inference in DBNs [Kja92] Unfortunately, in most cases, these algorithms also end up maintaining a belief state a distribution over most or all of the variables in a time slice. Furthermore, it can be shown [BK98] that the belief ....
....tree structure allows variable sized bins, and therefore greater flexibility in matching the number of parameters to the complexity of the distribution. 5 Experimental results To provide a more realistic comparison, we tested the different variants of our algorithm on the practical WATER DBN [JKOP89], used for monitoring the biological processes of a water purification plant. Comparable results were obtained for the CAPITAL network. The WATER DBN had a substantially larger state space, with 27,648 possible values taken by the (non evidence) variables. The structure of the WATER network is ....
F.V. Jensen, U. Kjærulff, K.G. Olesen, and J. Pedersen. An expert system for control of waste water treatment--- a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, Denmark, 1989.
Tractable Inference for Complex Stochastic Processes - Boyen (1998) (96 citations) (Correct)
....the applicability of our ideas in the context of a monitoring task, showing that orders of magnitude faster inference can be achieved with only a small degradation in accuracy. 1 Introduction The ability to model and reason about stochastic processes is fundamental to many applications [6, 8, 3, 14]. A number of formal models have been developed for describing situations of this type, including Hidden Markov Models [15] Kalman Filters [9] and Dynamic Bayesian Networks [4] These very different models all share the same underlying Markov assumption, the fact that the future is ....
....if we have two (sets of) variables that are highly correlated, splitting them into two separate subprocesses is not a good idea. Our experimental results illustrate these tradeoffs. 5. 3 Experimental results We validated this algorithm in the context of two real life DBNs: the WATER network [8], used for monitoring the biological processes of a water purification plant; and the BAT network [6] used for monitoring freeway traffic (see Figure 3) We added a few evidence nodes to WATER, which did not have any; these duplicate a few of the state variables SensorValid1 FYdotDiff1 ....
F.V. Jensen, U. Kjærulff, K.G. Olesen, and J. Pedersen. An expert system for control of waste water treatment---a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, 1989. In Danish.
Efficient Reinforcement Learning in Factored MDPs - Kearns, Koller (1999) (2 citations) (Correct)
No context found.
F.V. Jensen, U. Kjrulff, K.G. Olesen, and J. Pedersen. An expert system for control of waste water treatment---a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, 1989. In Danish.
Efficient Reinforcement Learning in Factored MDPs - Michael Kearns Att (1999) (2 citations) (Correct)
No context found.
F.V. Jensen, U. Kjrulff, K.G. Olesen, and J. Pedersen. An expert system for control of waste water treatment---a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, 1989. In Danish.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC