T. M. Rao

An evolutionary algorithmic approach to learning a Bayesian network from complete data

Discovering relationships between variables is crucial for interpreting data from large databases. Relationships between variables can be modeled using a Bayesian network. The challenge of learning a Bayesian network from a complete dataset grows exponentially with the number of variables in the database and the number of states in each variable. It therefore becomes important to identify promising heuristics for exploring the space of possible networks. This paper utilizes an evolutionary algorithmic approach, Particle Swarm Optimization (PSO) to perform this search. A fundamental problem with a search for a Bayesian network is that of handling cyclic networks, which are not allowed. This paper explores the PSO approach, handling cyclic networks in two different ways. Results of network extraction for the well-studied ALARM network are presented for PSO simulations where cycles are broken heuristically at each step of the optimization and where networks with cycles are allowed to exist as candidate solutions, but are assigned a poor fitness. The results of the two approaches are compared and it is found that allowing cyclic networks to exist in the particle swarm of candidate solutions can dramatically reduce the number of objective function evaluations required to converge to a target fitness value.

Snow-plow route planning using AI search

Snow plowing comprises a major portion of the total municipal budget in many communities. In these times of severe budget stress, it is vital to find ways to perform this essential service in an efficient manner. Optimizing the routes travelled by plows is one way to reduce costs. This problem, however, can be shown to be an NP-Hard problem at its core, with several additional complications. In this paper, we demonstrate how this problem can be formulated as a state-space search problem, and how one can employ AI techniques such as A-Star (A*) search to compute optimal or close-to-optimal routes. Specifically, we have developed A*-SnowPlowRouter (A*SPR), a Java software application that uses a “roads data set” of a town and generates efficient route plans. This data set is generally obtained from the municipal Geographical Information System (GIS). Our procedure transforms this data into a directed graph representation, augmented with road priority data. We then employ the A* technique to generate routes which minimize travel distance, avoid U-Turns, and reach higher-priority roads before lower-priority roads. The route plans computed by A*SPR are then exported into a format that can be displayed on the GIS. We have experimented with road networks in several local municipalities, and obtained routes that are significantly better than the ones currently used.

A Distributed evolutionary algorithmic approach to the least-cost connected constrained sub-graph and power control problem

When wireless sensors are capable of variable transmit power and are battery powered, it is important to select the appropriate transmit power level for the node. Lowering the transmit power of the sensor nodes imposes a natural clustering on the network and has been shown to improve throughput of the network. However, a common transmit power level is not appropriate for inhomogeneous networks. A possible fitness-based approach, motivated by an evolutionary optimization technique, Particle Swarm Optimization (PSO) is proposed and extended in a novel way to determine the appropriate transmit power of each sensor node. A distributed version of PSO is developed and explored using experimental fitness to achieve an approximation of least-cost connectivity.

Robust recruitment near the edge of chaos and an application to mine sweeping

The field of robotics is in rapid development. As robots become cheaper to build, new applications involving many robots systems can be envisioned. One reason for using many robots is to achieve robustness. Having many robots, however, does not ensure robustness. A control strategy and robot behaviors must be engineered to incorporate robustness into the system. Swarm intelligence based approaches are popular for developing optimal and robust control strategies for systems of robots. Here we analyze the behavior of a swarm of robots modeled after a swarm of ants, where tasks are spatially distributed in the environment and robots/ants are recruited through short-range recruitment. For ants that move probabilistically in response to the short range signal and who adjust their probabilities such that they are near a phase change boundary, or edge-of-chaos, in the mean field analysis of their motions, we find a significant improvement in the robustness of the system.

Evolutionary algorithms for the edge biconnectivity augmentation problem

A graph is edge-biconnected if it requires the removal of at least two edges to disconnect it. Assume that we have weighted graph that is not biconnected, and an additional set of augmentation edges. The (NP-hard) edge biconnectivity augmentation problem is to select a minimal subset of the augmentation edges, whose inclusion will cause the graph to be biconnected. This paper explores the application of particle swarm optimization and genetic algorithms for this problem.

Route allocation for multiple snowplows using genetic algorithms

The road network of a small town is represented as a directed graph where each road junction is a vertex and each road segment (which has a length and a priority value) is a directed edge. We assume that there are several plows available to service the roads. We seek to compute an optimal allocation of routes to plows. Each route begins and ends at the same (depot) vertex. The union of all plow routes must cover every edge in the graph at least once. In addition, we wish to minimize the following parameters: total distance covered by all plows (thereby minimizing the deadhead miles), amount of variation in the mileage covered by individual plows (thereby dividing the workload equitably), number of u-turns and extent of priority misplacements. In this paper, we propose a Genetic Algorithms (GA)-based solution to compute a near-optimal route allocation and the minimization of other parameters simultaneously. This algorithm is based on our GA solution to the 1-plow problem reported earlier. We have developed a Java application that implements our algorithm. Our experiments with reasonably large graphs have yielded good solutions. These solutions are especially useful in snowplow routing for small towns, as plowing costs consume significant portions of the total municipal budgets of these communities. Most of the route planning in small towns is currently done manually and routes have evolved over time by experience. In these times of severe budget stress, route allocation using our approach can help in performing this essential service in a more efficient manner.

Computing optimal snowplow route plans using Genetic Algorithms

The road network of a small town is represented by a directed graph. Road junctions are the vertices of this graph and each road segment (which has a length and a priority value) is represented by a directed edge. Priority values are numbers 1, 2, etc. with the assumption that 1 is the highest priority. We seek to compute an optimal route map that begins at a particular vertex (the depot) and covers all the edges at least once and returns to the start vertex. The parameters that we wish to minimize are: the total distance covered (thereby minimizing the deadhead miles), the number of u-turns and priority misplacements. In this paper, we propose a Genetic Algorithms-based solution to compute near-optimal route maps in such a graph. Specifically, we have developed a Java software application that generates route maps that minimize a linear combination of the three parameters. We have experimented with reasonably large graphs and obtained good solutions. These solutions are especially useful in snowplow routing for small towns, as plowing costs consume significant portions of the total municipal budgets of these communities. Most of the route planning is currently done manually and routes have evolved over time by experience. In these times of severe budget stress, route planning using our approach can help in performing this essential service in an efficient manner.