Kai Olav Ellefsen

K. Ege de Bruin, Kyrre Glette, Kai Olav Ellefsen et al.

Optimizing the body and brain of a robot is a coupled challenge: the morphology determines what control strategies are effective, while the control parameters influence how well the morphology performs. This joint optimization can be done through nested loops of evolutionary and learning processes, where the control parameters of each robot are learned independently. However, the control parameters learned by one robot may contain valuable information for others. Thus, we introduce a social learning approach in which robots can exploit optimized parameters from their peers to accelerate their own brain optimization. Within this framework, we systematically investigate how the selection of teachers, deciding which and how many robots to learn from, affects performance, experimenting with virtual soft robots in four tasks and environments. In particular, we study the effect of inheriting experience from morphologically similar robots due to the tightly coupled body and brain in robot optimization. Our results confirm the effectiveness of building on others' experience, as social learning clearly outperforms learning from scratch under equivalent computational budgets. In addition, while the optimal teacher selection strategy remains open, our findings suggest that incorporating knowledge from multiple teachers can yield more consistent and robust improvements.

K. Ege de Bruin, Kyrre Glette, Kai Olav Ellefsen

The co-optimization of a robot's body and brain presents a coupled challenge: the morphology constrains which control strategies are effective, while the control determines how well the morphology performs. To address this, we combine morphology optimization as evolution with controller optimization as lifetime learning, utilizing Lamarckian inheritance to transfer learned controller parameters from parent to offspring. In dynamic environments, existing literature presents conflicting evidence: while traditional evolutionary theory often suggests Lamarckian inheritance lacks benefit, recent studies in evolutionary robotics indicate it can improve performance. We hypothesize that this is because previous works have not included all relevant variables with dynamic environments. In this work, we show that the benefit of Lamarckian inheritance depends on two variables: how conflicting the environmental changes are to robot control, and the predictability of those changes for the robotic agent. Using virtual soft robots and two different learning approaches, Bayesian optimization and reinforcement learning, we show that Lamarckian inheritance only underperforms Darwinian inheritance when the changes are both conflicting and unpredictable. We find that adding a sensor to detect environmental changes restores the benefits for Lamarckian inheritance in conflicting environments, by allowing robotic agents to predict the need for a different behavior, thereby generalizing their control.

Samuel Valland Lyngset, Tor Viljen Raanaas, Gard Sveipe et al.

When fine-tuning Large Language Models (LLMs), there has been success in minimizing both memory usage and computation with Parameter-Efficient Fine-Tuning (PEFT), like Low Rank Adaptation (LoRA). In this article, we have explored whether this approach is transferable to the world of robotics and Reinforcement Learning (RL), allowing learning with reduced memory usage and improved computational performance. Specifically, we focused on a version of multi-task robotics, where a library of specialist policies are created. In such a library memory efficiency is especially important. We used a Proximal Policy Optimization (PPO) algorithm and fine-tuned a baseline model to different tasks using LoRA. Our results demonstrate that, depending on the hyperparameters, LoRA can minimize memory usage by a factor of 20-160 compared to full fine-tuning of all layers. This implies a 90-95% storage saving when deploying a library of many (10-50) specialized policies, which can be the differentiating factor between being able to store the entire library in memory or having to use swap-memory in an applied robotics setting. At the same time, our results indicate that there is no significant difference in the success-rate between full fine-tuning and LoRA fine-tuning for the selected tasks.

Ege de Bruin, Kyrre Glette, Kai Olav Ellefsen

It is well known that intra-life learning, defined as an additional controller optimization loop, is beneficial for evolving robot morphologies for locomotion. In this work, we investigate this further by comparing it in two different environments: an easy flat environment and a more challenging hills environment. We show that learning is significantly more beneficial in a hilly environment than in a flat environment and that it might be needed to evaluate robots in a more challenging environment to see the benefits of learning.

Jørgen Nordmoen, Frank Veenstra, Kai Olav Ellefsen et al.

In modular robotics, modules can be reconfigured to change the morphology of the robot, making it able to adapt for specific tasks. However, optimizing both the body and control is a difficult challenge due to the intricate relationship between fine-tuning control and morphological changes that can invalidate such optimizations. To solve this challenge we compare three different Evolutionary Algorithms on their capacity to optimize morphologies in modular robotics. We compare two objective-based search algorithms, with MAP-Elites. To understand the benefit of diversity we transition the evolved populations into two difficult environments to see if diversity can have an impact on solving complex environments. In addition, we analyse the genealogical ancestry to shed light on the notion of stepping stones as key to enable high performance. The results show that MAP-Elites is capable of evolving the highest performing solutions in addition to generating the largest morphological diversity. For the transition between environments the results show that MAP-Elites is better at regaining performance by promoting morphological diversity. With the analysis of genealogical ancestry we show that MAP-Elites produces more diverse and higher performing stepping stones than the other objective-based search algorithms. Transitioning the populations to more difficult environments show the utility of morphological diversity, while the analysis of stepping stones show a strong correlation between diversity of ancestry and maximum performance on the locomotion task. The paper shows the advantage of promoting diversity for solving a locomotion task in different environments for modular robotics. By showing that the quality and diversity of stepping stones in Evolutionary Algorithms is an important factor for overall performance we have opened up a new area of analysis and results.

Jørgen Nordmoen, Frank Veenstra, Kai Olav Ellefsen et al.

In Evolutionary Robotics a population of solutions is evolved to optimize robots that solve a given task. However, in traditional Evolutionary Algorithms, the population of solutions tends to converge to local optima when the problem is complex or the search space is large, a problem known as premature convergence. Quality Diversity algorithms try to overcome premature convergence by introducing additional measures that reward solutions for being different while not necessarily performing better. In this paper we compare a single objective Evolutionary Algorithm with two diversity promoting search algorithms; a Multi-Objective Evolutionary Algorithm and MAP-Elites a Quality Diversity algorithm, for the difficult problem of evolving control and morphology in modular robotics. We compare their ability to produce high performing solutions, in addition to analyze the evolved morphological diversity. The results show that all three search algorithms are capable of evolving high performing individuals. However, the Quality Diversity algorithm is better adept at filling all niches with high-performing solutions. This confirms that Quality Diversity algorithms are well suited for evolving modular robots and can be an important means of generating repertoires of high performing solutions that can be exploited both at design- and runtime.

Kai Olav Ellefsen, Jim Torresen

A long-standing challenge in Reinforcement Learning is enabling agents to learn a model of their environment which can be transferred to solve other problems in a world with the same underlying rules. One reason this is difficult is the challenge of learning accurate models of an environment. If such a model is inaccurate, the agent's plans and actions will likely be sub-optimal, and likely lead to the wrong outcomes. Recent progress in model-based reinforcement learning has improved the ability for agents to learn and use predictive models. In this paper, we extend a recent deep learning architecture which learns a predictive model of the environment that aims to predict only the value of a few key measurements, which are be indicative of an agent's performance. Predicting only a few measurements rather than the entire future state of an environment makes it more feasible to learn a valuable predictive model. We extend this predictive model with a small, evolving neural network that suggests the best goals to pursue in the current state. We demonstrate that this allows the predictive model to transfer to new scenarios where goals are different, and that the adaptive goals can even adjust agent behavior on-line, changing its strategy to fit the current context.

Kai Olav Ellefsen, Charles Patrick Martin, Jim Torresen

Gaining a better understanding of how and what machine learning systems learn is important to increase confidence in their decisions and catalyze further research. In this paper, we analyze the predictions made by a specific type of recurrent neural network, mixture density RNNs (MD-RNNs). These networks learn to model predictions as a combination of multiple Gaussian distributions, making them particularly interesting for problems where a sequence of inputs may lead to several distinct future possibilities. An example is learning internal models of an environment, where different events may or may not occur, but where the average over different events is not meaningful. By analyzing the predictions made by trained MD-RNNs, we find that their different Gaussian components have two complementary roles: 1) Separately modeling different stochastic events and 2) Separately modeling scenarios governed by different rules. These findings increase our understanding of what is learned by predictive MD-RNNs, and open up new research directions for further understanding how we can benefit from their self-organizing model decomposition.

Kai Olav Ellefsen, Herman A. Lepikson, Jan C. Albiez

An important open problem in robotic planning is the autonomous generation of 3D inspection paths -- that is, planning the best path to move a robot along in order to inspect a target structure. We recently suggested a new method for planning paths allowing the inspection of complex 3D structures, given a triangular mesh model of the structure. The method differs from previous approaches in its emphasis on generating and considering also plans that result in imperfect coverage of the inspection target. In many practical tasks, one would accept imperfections in coverage if this results in a substantially more energy efficient inspection path. The key idea is using a multiobjective evolutionary algorithm to optimize the energy usage and coverage of inspection plans simultaneously - and the result is a set of plans exploring the different ways to balance the two objectives. We here test our method on a set of inspection targets with large variation in size and complexity, and compare its performance with two state-of-the-art methods for complete coverage path planning. The results strengthen our confidence in the ability of our method to generate good inspection plans for different types of targets. The method's advantage is most clearly seen for real-world inspection targets, since traditional complete coverage methods have no good way of generating plans for structures with hidden parts. Multiobjective evolution, by optimizing energy usage and coverage together ensures a good balance between the two - both when 100% coverage is feasible, and when large parts of the object are hidden.

Joel Lehman, Jeff Clune, Dusan Misevic et al.

Biological evolution provides a creative fount of complex and subtle adaptations, often surprising the scientists who discover them. However, because evolution is an algorithmic process that transcends the substrate in which it occurs, evolution's creativity is not limited to nature. Indeed, many researchers in the field of digital evolution have observed their evolving algorithms and organisms subverting their intentions, exposing unrecognized bugs in their code, producing unexpected adaptations, or exhibiting outcomes uncannily convergent with ones in nature. Such stories routinely reveal creativity by evolution in these digital worlds, but they rarely fit into the standard scientific narrative. Instead they are often treated as mere obstacles to be overcome, rather than results that warrant study in their own right. The stories themselves are traded among researchers through oral tradition, but that mode of information transmission is inefficient and prone to error and outright loss. Moreover, the fact that these stories tend to be shared only among practitioners means that many natural scientists do not realize how interesting and lifelike digital organisms are and how natural their evolution can be. To our knowledge, no collection of such anecdotes has been published before. This paper is the crowd-sourced product of researchers in the fields of artificial life and evolutionary computation who have provided first-hand accounts of such cases. It thus serves as a written, fact-checked collection of scientifically important and even entertaining stories. In doing so we also present here substantial evidence that the existence and importance of evolutionary surprises extends beyond the natural world, and may indeed be a universal property of all complex evolving systems.

Charles P. Martin, Kai Olav Ellefsen, Jim Torresen

Musical performance requires prediction to operate instruments, to perform in groups and to improvise. In this paper, we investigate how a number of digital musical instruments (DMIs), including two of our own, have applied predictive machine learning models that assist users by predicting unknown states of musical processes. We characterise these predictions as focussed within a musical instrument, at the level of individual performers, and between members of an ensemble. These models can connect to existing frameworks for DMI design and have parallels in the cognitive predictions of human musicians. We discuss how recent advances in deep learning highlight the role of prediction in DMIs, by allowing data-driven predictive models with a long memory of past states. The systems we review are used to motivate musical use-cases where prediction is a necessary component, and to highlight a number of challenges for DMI designers seeking to apply deep predictive models in interactive music systems of the future.