Computer Science Faculty Research

Current faculty research interests center on robotics applications, the simulation of cognitive behaviors via artificial neural networks, and the use of alternative computing paradigms as problem solving techniques.

Robotics
This research attempts to demonstrate cognitive behaviors involving interaction between a simulated “brain” and the real world. The intention is for a device to perform meaningful actions that are significant in themselves—that is, without requiring interpretation by the experimenter. This line of research is important for a variety of reasons:

1. Conceptual and software-only models are fine, but until there is an actual interface with the physical universe it is all too easy to gloss over the real issues involved in trying to achieve any form of intelligent behavior in an artificial system.
2. New problems, concepts, perspectives, and solutions are consistently brought to one’s attention.
3. Other individuals (current and prospective students, faculty, funding agencies, etc.) are, in general, captivated more by physical demonstrations than by conceptual models.
4. A useful theory is not the same as a useful device and in many circumstances it is ultimately the device which not only validates the theory but popularizes it, as well.

Recent work has involved employing a digital camera as a robot eye for visual tracking, translation and scale invariant recognition, and face selection. Other projects have involved work on the learning of arm movements, examining contextual effects during serial learning, and creating a simulated neural network brain which was taught to react to various visual input prompts and respond accordingly, either by speaking (via a speech synthesis program) or by moving an arm (thus far simulated in software).

Near-term projects include attention-oriented eye (camera) movement, improved object recognition by a neural network, and purposeful arm movements.

Longer term projects will address issues such as face recognition, expanded speech synthesis capabilities, responsiveness to voice, and copy-cat motions in an artificial arm.

Advanced Cognitive Behaviors in an Artificial Neural Network
This research pertains to the development of an artificial neural network for multilevel interleaved and creative serial order cognitive behavior. This encompasses the design, development, and operation of a working model that provides for the acquisition, recall, and generation of temporally ordered information. These features are realized via a synergistic combination of subsystems that enable predictive learning, sequence interleaving, and new sequence creation. Together, these features provide a medium for demonstrating a variety of cognitive processes typical of intelligent behavior. Sequence learning is implemented as a Hebbian-based error-correcting paradigm that allows for rapid real-time training with minimum catastrophic interference and support for high-capacity sequence storage. Basic predictive learning is then used to illustrate common functions such as alphabet mastery, spelling, acquisition of mathematical facts, memorization of a script, basic motor skills, traditional associative memory, and an ability to form multiple associations with a single stimulus. Learned sequences can be interleaved to provide connectionist-based demonstrations of free association, transcription, route following, memory theatres, multiple trains of thought, complex motion, and a limited form of rehearsal. New sequences can then be crafted from previously learned sequences by using a generalized form of variable binding and can be used to help explicate such cognitive tasks as counting, solving mathematical expressions based on well-learned number facts, understanding simple pronoun referents in sentences, protolanguage reading comprehension, rule formation and the development of commonsense knowledge via inductive reasoning, the acquisition and deployment of external memory strategies, and a sophisticated nonstereotypical sequence-processing capability. By exhibiting such proficiency in a single temporally oriented network, this model takes a significant step toward the development of autonomous artificial systems capable of manifesting many of the characteristics that exemplify intelligence. In addition, the model embraces the search for common underlying cognitive principles and points to a number of promising areas for future research. The network is implemented in a customized simulation software system that permits a high degree of flexibility in model development and operation.

Future projects include the design of an asynchronous model (using purely distributed representations) with capabilities analogous to those evidenced in the existing model, integrating pattern learning, providing natural explanations/sources for sequence termination and creation processes, and exploring the role of expanded contextual influences.

(Click to magnify)

Student Research

Student research is the natural progression from a viable faculty research agenda. Not only does it provide valuable support for faculty research efforts, but it presents the challenges of the discipline in a unique way and offers a powerful opportunity for hands-on learning. All Computer Science students are encouraged to engage in research related activities during their time at Samford.

The department offers students a variety of research opportunities in the design and development of cognitive systems and robotics applications and in collaborative interdisciplinary projects (e.g., bioinformatics, neuroinformatics, cognitive science).

Students can obtain credit toward their major or minor by taking the established research course (COSC 410) for 1-4 hours of credit per semester. In addition, the Special Topics (COSC 460) and Senior Project (COSC 495) courses provide the chance for student project oriented research. Opportunities exist for students with innovative research projects to present their work at such forums as the National Conference on Undergraduate Research, the ACM MidSoutheast Regional Conference, the Alabama Academy of Science, and other discipline specific conferences.