Piero P. Bonissone’s Research Interests:
Knowledge Based Model Lifecycle
|
Timeline
|
|
|
Papers
|
|
(2010) [6] P. Bonissone, “Soft Computing: A Continuously Evolving
Concept”, Int. J. Computational
Intelligence Systems, to appear, 2010 - [GE GR Technical Report,
2009GRC845, Sept. 2009 (pdf)]. Soft
Computing (SC) is a concept with constantly evolving semantics, as
researchers have adopted its main philosophy while adding various
interpretations and facets to this concept.
Originally defined as a loose association or partnership of
components, SC has gone through several transformational phases. This paper will trace some of the phases
experienced by the author as part of his understanding of the evolution of SC
and its role in constructing decision-making models. The first phase is the hybridization phase,
driven by the inherit ease of integration of SC components. The second phase
is a two-level model characterization, based on the split between
object-level and meta-level reasoning. This phase, inspired by traditional AI
problem formulation, led to a third phase, in which we addressed the
knowledge and meta-knowledge representation required by each of these
reasoning levels using a linguistics analogy. The fourth phase is the
extension of the heuristics used at the meta-level, e.g. Metaheuristics (MH’s) from evolutionary
MH’s to other MH’s methods. The fifth
and last phase, further described in this paper, is the proposal for a strong
separation between offline MH’s (used for design and tuning) and online MH’s
(used for models selection or aggregation.) This last view suggests a broader
use of SC components, since it enables us to use hybrid SC techniques at each
of the MH’s levels as well as at the object level. Furthermore, this separation facilitates
the model lifecycle management, which is required to maintain the models
vitality and prevent their obsolescence over time. (2006)
[5] P. Bonissone, A. Varma, K. Aggour, and Feng Xue, “Design of local fuzzy
models using evolutionary algorithms", Computational Statistics and
Data Analysis, 51:398-416, 2006, (pdf) 2007 – [GE GR Technical Report, 2006GRC594, The application of local fuzzy models to determine
the remaining life of a unit in a fleet of vehicles is described. Instead of
developing individual models based on the track history of each unit or
developing a global model based on the collective track history of the fleet,
local fuzzy models are used based on clusters of peers—similar units with
comparable utilization and performance characteristics. A local fuzzy
performance model is created for each cluster of peers. This is combined with
an evolutionary framework to maintain the models. A process has been defined to generate a
collection of competing models, evaluate their performance in light of the
currently available data, refine the best models using evolutionary search,
and select the best one after a finite number of iterations. This process is
repeated periodically to automatically update and improve the overall model.
To illustrate this methodology an asset selection problem has been
identified: given a fleet of industrial vehicles (diesel electric
locomotives), select the best subset for mission-critical utilization. To
this end, the remaining life of each unit in the fleet is predicted. The fleet is then sorted using this
prediction and the highest ranked units are selected. A series of experiments
using data from locomotive operations was conducted and the results from an
initial validation exercise are presented. The approach of constructing local
predictive models using fuzzy similarity with neighboring points along
appropriate dimensions is not specific to any asset type and may be applied
to any problem where the premise of similarity along chosen attribute
dimensions implies similarity in predicted future behavior. (2005) [4] P.
Bonissone, A. Varma, K. Aggour, “An Evolutionary Process for Designing and
Maintaining a Fuzzy Instance-based Model (FIM)”, Proc. First Workshop of
Genetic Fuzzy Systems (GFS 2005), We illustrate the
typical life cycle of a fuzzy knowledge-based model, from its development,
testing, optimization, and deployment, through the maintenance of its
knowledge base. After a brief review of related work on a static problem, the
underwriting of insurance applications, we focus on the design,
implementation, and updating of a Fuzzy Instance-based Model (FIM) for
prediction and classification in a dynamic environment illustrated by a fleet selection problem. After
formalizing the FIM, we describe its evolutionary-based design process and
evaluate its performance. Finally, we
advocate the use of evolutionary search, intertwined with local search, to
further improve model life cycle. (2004) [3] A.
Patterson, P. Bonissone, and M. Pavese "Six Sigma Quality Applied
Throughout the Lifecycle of and Automated Decision System", Journal of Quality and Reliability
Engineering International, 21(3):275-292, April 2005 (pdf) – [GE GR Technical Report,
2003GRC365, Apr, 2004 (pdf)]. Automated decision-making systems have been deployed
in many industrial, commercial, and financial applications. The needs for
such systems are usually motivated by requirements for variation reduction,
capacity increase, cost and cycle time reduction, and end-to-end traceability
of the transaction or product. Before
we can use any automated decision-making system in a production environment
we must develop a strategy to insure high quality throughout its entire life
cycle. We need to guarantee its
performance through a rigorous Design for Six Sigma process (DFSS). This
process includes validation, tuning, and production testing of the system.
Once the system is in production we must monitor and maintain its performance
over its lifecycle. In this paper we
will outline the Six Sigma process that led to the
deployment of an automated decision-making system in one of the GE Financial
Assurance businesses. (2004) [2] P. Bonissone, “Development and
Maintenance of Fuzzy Models in Financial Applications”, in Soft
Methodology and Random Information Systems, Lopez-Diaz, Gil, Grzegorzewski, Hyrniewicz,
Lawry (Eds.), Springer, 2004 We
illustrate the typical life cycle of a fuzzy knowledge-based model, starting
from its development, testing, optimization, and deployment, and ending with
the maintenance of its knowledge base. We illustrate this process within the
context of an underwriting insurance application. First we define some key
concepts of soft computing models and discuss some design tradeoffs that must
be addressed. Then we focus on the design and implementation of a fuzzy
rule-based classifier (FRC). We establish a standard reference dataset (SRD),
consisting of 3,000 insurance applications with their corresponding
decisions. The SRD exemplifies the results achieved by an ideal, optimal
classifier, and represents the target for our design. We apply evolutionary
algorithms to perform an off-line optimization of the design parameters of
the classifier, modifying its behavior to approximate this target. The SRD is
also used as a reference for testing and performing a five-fold
cross-validation of the classifiers. Finally, we focus on the monitoring and
maintenance of the FRC. We describe a fusion architecture that supports an
off-line quality assurance process of the on-line FRC. The fusion module
takes the outputs of multiple classifiers, determines their degree of
consensus, and compares their overall agreement with the decision made by the
FRC. From this analysis, we can identify the most suitable cases to update
the SRD, to audit, or to be reviewed by senior underwriters. (2003) [1] P. Bonissone, “The life
cycle of a fuzzy knowledge-based classifier”, Proc. North American Fuzzy
Information Processing Society (NAFIPS 2003), pp. 488-494, The typical
lifecycle of a knowledge-based model starts from its development, testing,
optimization, and deployment, and continues with its maintenance phase. The latter consists of monitoring the
model’s performance, editing its knowledge base to prevent obsolescence, and
updating the model when required. Quite often, however, models are
handcrafted, i.e., a large amount of manual intervention is used in the
earlier phase of their lifecycle. This
leaves the maintenance phase as an “after-thought”, often requiring a similar
level of manual efforts. We propose a process, based on evolutionary
algorithms, that follows the model throughout its entire lifecycle. For deployment, it generates a collection
of competing models, evaluates their performance in light of the currently
available data, refines the best models using evolutionary search, and after
a finite number of iterations, generates the best-found model. This process
is repeated periodically to automatically produce new updated versions of the
model. We chose an asset selection
problem to illustrate this methodology. Given a fleet of industrial vehicles
(diesel electric locomotives), we want to select the best subset (of fixed or
variable size) for mission-critical utilization. To this end, we predict the
remaining life for each unit in the fleet. We then sort the fleet using this
prediction and select the highest ranked units. The model chosen to perform this
prediction/selection task is a fuzzy instance-based model. Unlike functional approximators that
require an off-line supervised learning stage to create a hyper-surface in
the cross-space state-output, instance-based models (IBM’s) only require
accessing the data in a local neighborhood of the new point defined by the
query. IBM’s rely on a collection of
previously experienced data stored in their raw representation. Unlike
Case-Based Reasoning (CBR), they do not need to be refined, abstracted and
organized as cases. Like CBR, IBM’s represent an analogical approach to
reasoning since they rely on previous instances of similar problems and use
them to create an ensemble of local models. Hence the definition of
similarity plays a critical role in their performance. Typically, similarity
will be a dynamic concept and will change over the use of the IBM’s. Therefore, it is important to apply
learning methodologies to define and adapt it. Furthermore, the concept of
similarity is not crisply defined, creating the need to allow for some degree
of vagueness in its evaluation.
Hence, we propose the use of Fuzzy IBM’s (F-IBM’s). We address the issue of similarity by
evolving the design of a similarity function in conjunction with the design
of the attribute space in which the similarity is to be evaluated. Specifically, we use four steps: (1) Retrieval of similar instances from the database (DB); (2) Evaluation of
similarity measures between the probe and the retrieved instances; (3)
Creation of local models using the most similar instances (weighted by
their similarity measures); (4) Aggregation
of outputs of local models to
probe. Within the example of asset selection, we show the accuracy of
the evolved F-IBM’s, their robustness to information loss, and the benefit of
their automated updating process to avoid performance loss. Finally, we
advocate the use of evolutionary search intertwined with local search to
further improve model life cycle. |
Patents (Issued / Applications)
|
|
(2009) Method and system for
forecasting reliability of assets, P. Bonissone, K. Aggour,
A. Varma, US Patent No. 7,509,235, (March 24, 2009). (2003) System for optimization of insurance
underwriting suitable for use by an automated system, P. Bonissone, R. Messmer, A. Patterson, D.
Yang, M. Pavese, R. Subbu, K. Aggour, PUB_20030187702 filed (2003) Process for optimization of
insurance underwriting suitable for use by an automated system, P. Bonissone, R. Messmer, A. Patterson, D.
Yang, M. Pavese, R. Subbu, K. Aggour, PUB_20030187701 filed |
Supervised MS –
PhD Theses
|
|
- |
Projects
|
|Bonissone
Home Page| GE Research
Computer and Decision Sciences
|
|
