Wayne Liu - Academia.edu (original) (raw)
Uploads
Papers by Wayne Liu
World Wide Web, 1998
This paper addresses the problem of learning Bayesian network structures from data by using an in... more This paper addresses the problem of learning Bayesian network structures from data by using an information theoretic dependency analysis approach. Based on our three-phase construction mechanism, two efficient algorithms have been developed. One of our algorithms deals with a special case where the node ordering is given, the algorithm only require ) ( 2 N O CI tests and is correct given that the underlying model is DAG-Faithful . The other algorithm deals with the general case and requires ) ( 4 N O conditional independence (CI) tests. It is correct given that the underlying model is monotone DAG-Faithful (see Section 4.4). A system based on these algorithms has been developed and distributed through the Internet. The empirical results show that our approach is efficient and reliable. Definition 2.6. For a probabilistic model M, a graph G is a dependency map (D-map) of M if every independence relationship of M can be expressed in G. A graph without any edges is a trivial D-map since every node is independent to all other nodes. Definition 2.7. For a probabilistic model M, a graph G is an independency map (I-map) of M if every independence relationship derived from G is true in M. A complete graph where every node is connected with every other node is a trivial I-map since it contains no independence relationships. Definition 2.8. A graph G is a minimum I-map of a model M if it is an I-map of M and the removal of any arc from G yields a graph that is not an I-map of M. Definition 2.9. If a graph G is both a D-map and an I-map of M, we call it a perfect map (P-map) of M.
International Journal of Heat and Mass Transfer, 1995
The behavior of a molten metal droplet impinging, spreading and solidifying on a solid substrate ... more The behavior of a molten metal droplet impinging, spreading and solidifying on a solid substrate is relevant to manufacturing processes such as splat cooling and spray deposition. In this study, we have conducted experiments aimed at the investigation of the solidification and cooling of a metal droplet after impact. Temperature measurements were conducted during and after solidification. Some results are presented to illustrate the effect of superheat and substrate material on the cooling rate. We have also estimated the thermal contact coefficient between splat and substrate by matching for a number of conditions the experimental data to predictions of a heat transfer and phase change model. The results suggest that this coefficient can decrease by an order of magnitude during solidification for the case of a splat on metal substrates. For splats on quartz, where there is good bonding, the thermal contact coefficient appears to stay the same before and after solidification.
World Wide Web, 1998
This paper addresses the problem of learning Bayesian network structures from data by using an in... more This paper addresses the problem of learning Bayesian network structures from data by using an information theoretic dependency analysis approach. Based on our three-phase construction mechanism, two efficient algorithms have been developed. One of our algorithms deals with a special case where the node ordering is given, the algorithm only require ) ( 2 N O CI tests and is correct given that the underlying model is DAG-Faithful . The other algorithm deals with the general case and requires ) ( 4 N O conditional independence (CI) tests. It is correct given that the underlying model is monotone DAG-Faithful (see Section 4.4). A system based on these algorithms has been developed and distributed through the Internet. The empirical results show that our approach is efficient and reliable. Definition 2.6. For a probabilistic model M, a graph G is a dependency map (D-map) of M if every independence relationship of M can be expressed in G. A graph without any edges is a trivial D-map since every node is independent to all other nodes. Definition 2.7. For a probabilistic model M, a graph G is an independency map (I-map) of M if every independence relationship derived from G is true in M. A complete graph where every node is connected with every other node is a trivial I-map since it contains no independence relationships. Definition 2.8. A graph G is a minimum I-map of a model M if it is an I-map of M and the removal of any arc from G yields a graph that is not an I-map of M. Definition 2.9. If a graph G is both a D-map and an I-map of M, we call it a perfect map (P-map) of M.
International Journal of Heat and Mass Transfer, 1995
The behavior of a molten metal droplet impinging, spreading and solidifying on a solid substrate ... more The behavior of a molten metal droplet impinging, spreading and solidifying on a solid substrate is relevant to manufacturing processes such as splat cooling and spray deposition. In this study, we have conducted experiments aimed at the investigation of the solidification and cooling of a metal droplet after impact. Temperature measurements were conducted during and after solidification. Some results are presented to illustrate the effect of superheat and substrate material on the cooling rate. We have also estimated the thermal contact coefficient between splat and substrate by matching for a number of conditions the experimental data to predictions of a heat transfer and phase change model. The results suggest that this coefficient can decrease by an order of magnitude during solidification for the case of a splat on metal substrates. For splats on quartz, where there is good bonding, the thermal contact coefficient appears to stay the same before and after solidification.