Jan M. Zytkow

AAAI Conference 1996 Conference Paper

Incremental Discovery of Hidden Structure: Applications in Theory of Elementary Particles

• Jan M. Zytkow

Discovering hidden structure is a challenging, universal research task in Physics, Chemistry, Biology, and other disciplines. Not only must the elements of hidden structure be postulated by the discoverer, but they can only be verified by indirect evidence, at the level of observable objects. In this paper we describe a framework for hidden structure discovery, built on a constructive definition of hidden structure. This definition leads to operators that build models of hidden structure step by step, postulating hidden objects, their combinations and properties, reactions described in terms of hidden objects, and mapping between the hidden and the observed structure. We introduce the operator dependency diagram, which shows the order of operator application and model evaluation. Different observational knowledge supports different evaluation criteria, which lead to different search systems with verifiable sequences of operator applications. Isomorphfree structure generation is another issue critical for efficiency of search. We apply our framework in the system GELL-MANN, that hypothesizes hidden structure for elementary particles and we present the results of a large scale search for quark models.

AAAI Conference 1992 Conference Paper

Operational Definition Refinement: A Discovery Process

• Jan M. Zytkow

Operational definitions link scientific attributes to experimental situations, prescribing for the experimenter the actions and measurements needed to measure or control attribute values. While very important in real science, operational procedures have been neglected in machine discovery. We argue that in the preparatory stage of the empirical discovery process each operational definition must be adjusted to the experimental task at hand. This is done in the interest of error reduction and repeatability of measurements. Both small error and high repeatability are instrumental in theory formation. We demonstrate that operational procedure refinement is a discovery process that resembles the discovery of scientific laws. We demonstrate how the discovery task can be reduced to an application of the FAHRENHEIT discovery system. A new type of independent variables, the experiment refinement variables, have been introduced to make the application of FAHRENHEIT theoretically valid. This new extension to FAHRENHEIT uses simple operational procedures, as well as the system’s experimentation and theory formation capabilities to collect real data in a science laboratory and to build theories of error and repeatability that are used to refine the operational procedures. We present the application of FAHRENHEIT in the context of dispensing liquids in a chemistry laboratory.

AAAI Conference 1990 Conference Paper

Automated Discovery in a Chemistry Laboratory

• Jan M. Zytkow

AIJ Journal 1989 Journal Article

Data-driven approaches to empirical discovery

• Pat Langley
• Jan M. Zytkow

In this paper we track the development of research in empirical discovery. We focus on four machine discovery systems that share a number of features: the use of data-driven heuristics to constrain the search for numeric laws; a reliance on theoretical terms; and the recursive application of a few general discovery methods. We examine each system in light of the innovations it introduced over its predecessors, providing some insight into the conceptual progress that has occurred in machine discovery. Finally, we reexamine this research from the perspectives of the history and philosophy of science.

IJCAI Conference 1983 Conference Paper

Three Facets of Scientific Discovery

• Pat Langley
• Jan M. Zytkow
• Gary L. Bradshaw
• Herbert A. Simon

Scientific discovery is a complex process, and in this paper we consider three of its many facets - discovering laws of qualitative structure, finding quantitative relations between variables, and formulating sfructural models of reactions. We describe three discovery systems - GLAUBER, BACON, and DALTON - thr. t address these three aspects of the scientific process. GLAUBER forms classes of objects based on regularities in qualitative data, and states abstract laws in terms of these classes. BACON includes heuristics for finding numerical laws, for postulating intrinsic properties, and for noting common divisors. DALTON formulates molecular models that account for observed reactions, taking advantage of theoretical assumptions to direct its search if they are available. We show how each of the programs is capable of rediscovering laws or models that were found in the early days of chemistry. Finally, we consider some possble interactions between these systems, and the need for an integrated theory of discovery.