Paul A. Szerlip

Confronting the Challenge of Quality Diversity

In contrast to the conventional role of evolution in evolutionary computation (EC) as an optimization algorithm, a new class of evolutionary algorithms has emerged in recent years that instead aim to accumulate as diverse a collection of discoveries as possible, yet where each variant in the collection is as fit as it can be. Often applied in both neuroevolution and morphological evolution, these new quality diversity (QD) algorithms are particularly well-suited to evolutions inherent strengths, thereby offering a promising niche for EC within the broader field of machine learning. However, because QD algorithms are so new, until now no comprehensive study has yet attempted to systematically elucidate their relative strengths and weaknesses under different conditions. Taking a first step in this direction, this paper introduces a new benchmark domain designed specifically to compare and contrast QD algorithms. It then shows how the degree of alignment between the measure of quality and the behavior characterization (which is an essential component of all QD algorithms to date) impacts the ultimate performance of different such algorithms. The hope is that this initial study will help to stimulate interest in QD and begin to unify the disparate ideas in the area.

Interactively evolving harmonies through functional scaffolding

While the real-time focus of todays automated accompaniment generators can benefit instrumentalists and vocalists in their practice, improvisation, or performance, an opportunity remains specifically to assist novice composers. This paper introduces a novel such approach based on evolutionary computation called functional scaffolding for musical composition (FSMC), which helps the user explore potential accompaniments for existing musical pieces, or scaffolds. The key idea is to produce accompaniment as a function of the scaffold, thereby inheriting from its inherent style and texture. To implement this idea, accompaniments are represented by a special type of neural network called a compositional pattern producing network (CPPN), which produces harmonies by elaborating on and exploiting regularities in pitches and rhythms found in the scaffold. This paper focuses on how inexperienced composers can personalize accompaniments by first choosing any MIDI scaffold, then selecting which parts (e.g. the piano, guitar, or bass guitar) the CPPN can hear, and finally customizing and refining the computer-generated accompaniment through an interactive process of selection and mutation of CPPNs called interactive evolutionary computation (IEC). The potential of this approach is demonstrated by following the evolution of a specific accompaniment and studying whether listeners appreciate the results.

Indirectly Encoded Sodarace for Artificial Life

The aim of this paper is to introduce a lightweight twodimensional domain for evolving diverse and interesting artificial creatures. The hope is that this domain will fill a need for such an easily-accessible option for researchers who wish to focus more on the evolutionary dynamics of artificial life scenarios than on building simulators and creature encodings. The proposed domain is inspired by Sodarace, a construction set for two-dimensional creatures made of masses and springs. However, unlike the original Sodarace, the indirectly encoded Sodarace (IESoR) system introduced in this paper allows evolution to discover a wide range of complex and regular ambulating creature morphologies by encoding them with compositional pattern producing networks (CPPNs), which are an established indirect encoding originally introduced for encoding large-scale neural networks. The result, demonstrated through a technique called novelty search with local competition (which are combined through multiobjective search), is that IESoR can discover a wide breadth of interesting and functional creatures, suggesting its potential utility for future experiments in artificial life.

Functional Scaffolding for Composing Additional Musical Voices

Many tools for computer-assisted composition contain built-in music-theoretical assumptions that may constrain the output to particular styles. In contrast, this article presents a new musical representation that contains almost no built-in knowledge, but that allows even musically untrained users to generate polyphonic textures that are derived from the users own initial compositions. This representation, called functional scaffolding for musical composition (FSMC), exploits a simple yet powerful property of multipart compositions: The pattern of notes and rhythms in different instrumental parts of the same song are functionally related. That is, in principle, one part can be expressed as a function of another. Music in FSMC is represented accordingly as a functional relationship between an existing human composition, or scaffold, and a generated set of one or more additional musical voices. A human user without any musical expertise can then explore how the generated voice (or voices) should relate to the scaffold through an interactive evolutionary process akin to animal breeding. By inheriting from the intrinsic style and texture of the piece provided by the user, this approach can generate additional voices for potentially any style of music without the need for extensive musical expertise.

Association between preoperative activity level and functional outcome at 12 months following surgical decompression for lumbar spinal stenosis

OBJECTIVE This study defines the association of preoperative physical activity level with functional outcomes at 3 and 12 months following surgical decompression for lumbar spinal stenosis. METHODS Data were collected as a prospective observational registry at a single institution from 2012 through 2015, and then analyzed with a retrospective cohort design. Patients who were able to participate in activities outside the home preoperatively were compared to patients who did not participate in such activities, with respect to 3-month and 12-month functional outcomes postintervention, adjusted for relevant confounders. RESULTS Ninety-nine patients were included. At baseline, sedentary/inactive patients (n = 55) reported greater back pain, lower quality of life, and higher disability than similarly treated patients who were active preoperatively. Both cohorts experienced significant improvement from baseline in back pain, leg pain, disability, and quality of life at both 3 and 12 months after lumbar decompression surgery. At 3 months postintervention, sedentary/inactive patients reported more leg pain and worse disability than patients who performed activities outside the home preoperatively. However, at 12 months postintervention, there were no statistically significant differences between the two cohorts in back pain, leg pain, quality of life, or disability. Multivariate analysis revealed that sedentary/inactive patients had improved disability and higher quality of life after surgery compared to baseline. Active patients experienced greater overall improvement in disability compared to inactive patients. CONCLUSIONS Sedentary/inactive patients have a more protracted recovery after lumbar decompression surgery for spinal stenosis, but at 12 months postintervention can expect to reach similar long-term outcomes as patients who are active/perform activities outside the home preoperatively.

Indirectly encoding running and jumping sodarace creatures for artificial life

This article presents a lightweight platform for evolving two-dimensional artificial creatures. The aim of providing such a platform is to reduce the barrier to entry for researchers interested in evolving creatures for artificial life experiments. In effect the novel platform, which is inspired by the Sodarace construction set, makes it easy to set up creative scenarios that test the abilities of Sodarace-like creatures made of masses and springs. In this way it allows the researcher to focus on evolutionary algorithms and dynamics. The new indirectly encoded Sodarace (IESoR) system introduced in this article extends the original Sodarace by enabling the evolution of significantly more complex and regular creature morphologies. These morphologies are themselves encoded by compositional pattern-producing networks (CPPNs), an indirect encoding previously shown effective at encoding regularities and symmetries in structure. The capability of this lightweight system to facilitate research in artificial life is then demonstrated through both walking and jumping domains, in which IESoR discovers a wide breadth of strategies through novelty search with local competition.