A Two-Stage Online Approach for Collaborative Multi-Agent Planning under Uncertainty

In a team of multiple agents, the pursuance of a common goal is a defining characteristic. Since agents may have different capabilities, and effects of actions may be uncertain, a common goal can generally only be achieved through a careful cooperation between the different agents. In this work, we propose a novel two-stage planner that combines online planning at both team level and individual level through a subgoal delegation scheme. The proposal brings the advantages of online planning approaches to the multi-agent setting. A number of modifications are made to a classical UCT approximate algorithm to (i) adapt it to the application domains considered, (ii) reduce the branching factor in the underlying search process, and (iii) effectively manage uncertain information of action effects by using information fusion mechanisms. The proposed online multi-agent planner reduces the cost of planning and decreases the temporal cost of reaching a goal, while significantly increasing the chance of success of achieving the common goal. 

Directed Selection using Reinforcement Learning for the Examination Timetabling Problem

Traditional heuristic approaches to the Examination Timetabling Problem normally utilize a stochastic method during Optimization for the selection of the next examination to be considered for timetabling within the neighbourhood search process. This paper presents a technique whereby the stochastic method has been augmented with information from a weighted list gathered during the initial adaptive construction phase, with the purpose of intelligently directing examination selection. In addition, a Reinforcement Learning technique has been adapted to identify the most effective portions of the weighted list in terms of facilitating the greatest potential for overall solution improvement. The technique is tested against the 2007 International Timetabling Competition datasets with solutions generated within a time frame specified by the competition organizers. The results generated are better than those of the competition winner in seven of the twelve examinations, while being competitive for the remaining five examinations. This paper also shows experimentally how using reinforcement learning has improved upon our previous technique.

Glycosylation is an Androgen-Regulated Process Essential for Prostate Cancer Cell Viability

Steroid androgen hormones play a key role in the progression and treatment of prostate cancer, with androgen deprivation therapy being the first-line treatment used to control cancer growth. Here we apply a novel search strategy to identify androgen-regulated cellular pathways that may be clinically important in prostate cancer. Using RNASeq data, we searched for genes that showed reciprocal changes in expression in response to acute androgen stimulation in culture, and androgen deprivation in patients with prostate cancer. Amongst 700 genes displaying reciprocal expression patterns we observed a significant enrichment in the cellular process glycosylation. Of 31 reciprocally-regulated glycosylation enzymes, a set of 8 (GALNT7, ST6GalNAc1, GCNT1, UAP1, PGM3, CSGALNACT1, ST6GAL1 and EDEM3) were significantly up-regulated in clinical prostate carcinoma. Androgen exposure stimulated synthesis of glycan structures downstream of this core set of regulated enzymes including sialyl-Tn (sTn), sialyl Lewis(X) (SLe(X)), O-GlcNAc and chondroitin sulphate, suggesting androgen regulation of the core set of enzymes controls key steps in glycan synthesis. Expression of each of these enzymes also contributed to prostate cancer cell viability. This study identifies glycosylation as a global target for androgen control, and suggests loss of specific glycosylation enzymes might contribute to tumour regression following androgen depletion therapy.