Arjun Chandra

Evolving hybrid ensembles of learning machines for better generalisation

Ensembles of learning machines have been formally and empirically shown to outperform (generalise better than) single predictors in many cases. Evidence suggests that ensembles generalise better when they constitute members which form a diverse and accurate set. Additionally, there have been a multitude of theories on how one can enforce diversity within a combined predictor setup. We recently attempted to integrate these theories together into a co-evolutionary framework with a view to synthesising new evolutionary ensemble learning algorithms using the fact that multi-objective evolutionary optimisation is a formidable ensemble construction technique. This paper explicates on the intricacies of the proposed framework in addition to presenting detailed empirical results and comparisons with a wide range of algorithms in the machine learning literature. The framework treats diversity and accuracy as evolutionary pressures which are exerted at multiple levels of abstraction and is shown to be effective.

Ensemble Learning Using Multi-Objective Evolutionary Algorithms

Multi-objective evolutionary algorithms for the construction of neural ensembles is a relatively new area of research. We recently proposed an ensemble learning algorithm called DIVACE (DIVerse and ACcurate Ensemble learning algorithm). It was shown that DIVACE tries to find an optimal trade-off between diversity and accuracy as it searches for an ensemble for some particular pattern recognition task by treating these two objectives explicitly separately. A detailed discussion of DIVACE together with further experimental studies form the essence of this paper. A new diversity measure which we call Pairwise Failure Crediting (PFC) is proposed. This measure forms one of the two evolutionary pressures being exerted explicitly in DIVACE. Experiments with this diversity measure as well as comparisons with previously studied approaches are hence considered. Detailed analysis of the results show that DIVACE, as a concept, has promise.

DIVACE: Diverse and Accurate Ensemble Learning Algorithm

In order for a neural network ensemble to generalise properly, two factors are considered vital. One is the diversity and the other is the accuracy of the networks that comprise the ensemble. There exists a tradeoff as to what should be the optimal measures of diversity and accuracy. The aim of this paper is to address this issue. We propose the DIVACE algorithm which tries to produce an ensemble as it searches for the optimum point on the diversity-accuracy curve. The DIVACE algorithm formulates the ensemble learning problem as a multi-objective problem explicitly.

A Survey of Self-Awareness and Its Application in Computing Systems

Novel computing systems are increasingly being composed of large numbers of heterogeneous components, each with potentially different goals or local perspectives, and connected in networks which change over time. Management of such systems quickly becomes infeasible for humans. As such, future computing systems should be able to achieve advanced levels of autonomous behaviour. In this context, the systems ability to be self-aware and be able to self-express becomes important. This paper surveys definitions and current understanding of self-awareness and self-expression in biology and cognitive science. Subsequently, previous efforts to apply these concepts to computing systems are described. This has enabled the development of novel working definitions for self-awareness and self-expression within the context of computing systems.

Trade-Off Between Diversity and Accuracy in Ensemble Generation

Summary. Ensembles of learning machines have been formally and empirically shown to outperform (generalise better than) single learners in many cases. Evidence suggests that ensembles generalise better when they constitute members which form a diverse and accurate set. Diversity and accuracy are hence two factors that should be taken care of while designing ensembles in order for them to generalise better. There exists a trade-off between diversity and accuracy. Multi-objective evolutionary algorithms can be employed to tackle this issue to good effect. This chapter includes a brief overview of ensemble learning in general and presents a critique on the utility of multi-objective evolutionary algorithms for their design. Theoretical aspects of a committee of learners viz. the bias-variance-covariance decomposition and ambiguity decomposition are further discussed in order to support the importance of having both diversity and accuracy in ensembles. Some recent work and experimental results, considering classification tasks in particular, based on multi-objective learning of ensembles are then presented as we examine ensemble formation using neural networks and kernel machines.

Architectural Aspects of Self-Aware and Self-Expressive Computing Systems: From Psychology to Engineering

Work on human self-awareness is the basis for a framework to develop computational systems that can adaptively manage complex dynamic tradeoffs at runtime. An architectural case study in cloud computing illustrates the frameworks potential benefits.

EPiCS: Engineering Proprioception in Computing Systems

Modern compute systems continue to evolve towards increasingly complex, heterogeneous and distributed architectures. At the same time, functionality and performance are no longer the only aspects when developing applications for such systems, and additional concerns such as flexibility, power efficiency, resource usage, reliability and cost are becoming increasingly important. This does not only raise the question of how to efficiently develop applications for such systems, but also how to cope with dynamic changes in the application behaviour or the system environment. The EPiCS Project aims to address these aspects through exploring self-awareness and self-expression. Self-awareness allows systems and applications to gather and maintain information about their current state and environment, and reason about their behaviour. Self-expression enables systems to adapt their behaviour autonomously to changing conditions. Innovations in EPiCS are based on systematic integration of research in concepts and foundations, customisable hardware/software platforms and operating systems, and self-aware networking and middleware infrastructure. The developed technologies are validated in three application domains: computational finance, distributed smart cameras and interactive mobile media systems.

Learning to be Different: Heterogeneity and Efficiency in Distributed Smart Camera Networks

In this paper we study the self-organising behaviour of smart camera networks which use market-based handover of object tracking responsibilities to achieve an efficient allocation of objects to cameras. Specifically, we compare previously known homogeneous configurations, when all cameras use the same marketing strategy, with heterogeneous configurations, when each camera makes use of its own, possibly different marketing strategy. Our first contribution is to establish that such heterogeneity of marketing strategies can lead to system wide outcomes which are Pareto superior when compared to those possible in homogeneous configurations. However, since the particular configuration required to lead to Pareto efficiency in a given scenario will not be known in advance, our second contribution is to show how online learning of marketing strategies at the individual camera level can lead to high performing heterogeneous configurations from the system point of view, extending the Pareto front when compared to the homogeneous case. Our third contribution is to show that in many cases, the dynamic behaviour resulting from online learning leads to global outcomes which extend the Pareto front even when compared to static heterogeneous configurations. Our evaluation considers results obtained from an open source simulation package as well as data from a network of real cameras.

Static, Dynamic, and Adaptive Heterogeneity in Distributed Smart Camera Networks

We study heterogeneity among nodes in self-organizing smart camera networks, which use strategies based on social and economic knowledge to target communication activity efficiently. We compare homogeneous configurations, when cameras use the same strategy, with heterogeneous configurations, when cameras use different strategies. Our first contribution is to establish that static heterogeneity leads to new outcomes that are more efficient than those possible with homogeneity. Next, two forms of dynamic heterogeneity are investigated: nonadaptive mixed strategies and adaptive strategies, which learn online. Our second contribution is to show that mixed strategies offer Pareto efficiency consistently comparable with the most efficient static heterogeneous configurations. Since the particular configuration required for high Pareto efficiency in a scenario will not be known in advance, our third contribution is to show how decentralized online learning can lead to more efficient outcomes than the homogeneous case. In some cases, outcomes from online learning were more efficient than all other evaluated configuration types. Our fourth contribution is to show that online learning typically leads to outcomes more evenly spread over the objective space. Our results provide insight into the relationship between static, dynamic, and adaptive heterogeneity, suggesting that all have a key role in achieving efficient self-organization.

A novel adaptive weight selection algorithm for multi-objective multi-agent reinforcement learning

To solve multi-objective problems, multiple reward signals are often scalarized into a single value and further processed using established single-objective problem solving techniques. While the field of multi-objective optimization has made many advances in applying scalarization techniques to obtain good solution trade-offs, the utility of applying these techniques in the multi-objective multi-agent learning domain has not yet been thoroughly investigated. Agents learn the value of their decisions by linearly scalarizing their reward signals at the local level, while acceptable system wide behaviour results. However, the non-linear relationship between weighting parameters of the scalarization function and the learned policy makes the discovery of system wide trade-offs time consuming. Our first contribution is a thorough analysis of well known scalarization schemes within the multi-objective multi-agent reinforcement learning setup. The analysed approaches intelligently explore the weight-space in order to find a wider range of system trade-offs. In our second contribution, we propose a novel adaptive weight algorithm which interacts with the underlying local multi-objective solvers and allows for a better coverage of the Pareto front. Our third contribution is the experimental validation of our approach by learning bi-objective policies in self-organising smart camera networks. We note that our algorithm (i) explores the objective space faster on many problem instances, (ii) obtained solutions that exhibit a larger hypervolume, while (iii) acquiring a greater spread in the objective space.