From real-time adaptation to social learning in robot ecosystems.

While evolutionary robotics can create novel morphologies and controllers that are well-adapted to their environments, learning is still the most efficient way to adapt to changes that occur on shorter time scales. Learning proposals for evolving robots to date have focused on new individuals either learning a controller from scratch, or building on the experience of direct ancestors and/or robots with similar configurations. Here we propose and demonstrate a novel means for social learning of gait patterns, based on sensorimotor synchronization. Using movement patterns of other robots as input can drive nonlinear decentralized controllers such as CPGs into new limit cycles, hence encouraging diversity of movement patterns. Stable autonomous controllers can then be locked in, which we demonstrate using a quasi-Hebbian feedback scheme. We propose that in an ecosystem of robots evolving in a heterogeneous environment, such a scheme may allow for the emergence of generalist task-solvers from a population of specialists.

Central pattern generators evolved for real-time adaptation to rhythmic stimuli.

For a robot to be both autonomous and collaborative requires the ability to adapt its movement to a variety of external stimuli, whether these come from humans or other robots. Typically, legged robots have oscillation periods explicitly defined as a control parameter, limiting the adaptability of walking gaits. Here we demonstrate a virtual quadruped robot employing a bio-inspired central pattern generator (CPG) that can spontaneously synchronize its movement to a range of rhythmic stimuli. Multi-objective evolutionary algorithms were used to optimize the variation of movement speed and direction as a function of the brain stem drive and the centre of mass control respectively. This was followed by optimization of an additional layer of neurons that filters fluctuating inputs. As a result, a range of CPGs were able to adjust their gait pattern and/or frequency to match the input period. We show how this can be used to facilitate coordinated movement despite differences in morphology, as well as to learn new movement patterns.

Quantifying evoked responses through information-theoretical measures.

Information theory is a viable candidate to advance our understanding of how the brain processes information generated in the internal or external environment. With its universal applicability, information theory enables the analysis of complex data sets, is free of requirements about the data structure, and can help infer the underlying brain mechanisms. Information-theoretical metrics such as Entropy or Mutual Information have been highly beneficial for analyzing neurophysiological recordings. However, a direct comparison of the performance of these methods with well-established metrics, such as the t-test, is rare. Here, such a comparison is carried out by evaluating the novel method of Encoded Information with Mutual Information, Gaussian Copula Mutual Information, Neural Frequency Tagging, and t-test. We do so by applying each method to event-related potentials and event-related activity in different frequency bands originating from intracranial electroencephalography recordings of humans and marmoset monkeys. Encoded Information is a novel procedure that assesses the similarity of brain responses across experimental conditions by compressing the respective signals. Such an information-based encoding is attractive whenever one is interested in detecting where in the brain condition effects are present.

Complexity-based Encoded Information Quantification in Neurophysiological Recordings.

Brain activity differs vastly between sleep, cognitive tasks, and action. Information theory is an appropriate concept to analytically quantify these brain states. Based on neurophysiological recordings, this concept can handle complex data sets, is free of any requirements about the data structure, and can infer the present underlying brain mechanisms. Specifically, by utilizing algorithmic information theory, it is possible to estimate the absolute information contained in brain responses. While current approaches that apply this theory to neurophysiological recordings can discriminate between different brain states, they are limited in directly quantifying the degree of similarity or encoded information between brain responses. Here, we propose a method grounded in algorithmic information theory that affords direct statements about responses' similarity by estimating the encoded information through a compression-based scheme. We validated this method by applying it to both synthetic and real neurophysiological data and compared its efficiency to the mutual information measure. This proposed procedure is especially suited for task paradigms contrasting different event types because it can precisely quantify the similarity of neuronal responses.

MAP-Elites Enables Powerful Stepping Stones and Diversity for Modular Robotics.

In modular robotics modules can be reconfigured to change the morphology of the robot, making it able to adapt to specific tasks. However, optimizing both the body and control of such robots is a difficult challenge due to the intricate relationship between fine-tuning control and morphological changes that can invalidate such optimizations. These challenges can trap many optimization algorithms in local optima, halting progress towards better solutions. To solve this challenge we compare three different Evolutionary Algorithms on their capacity to optimize high performing and diverse morphologies and controllers in modular robotics. We compare two objective-based search algorithms, with and without a diversity promoting objective, with a Quality Diversity algorithm-MAP-Elites. The results show that MAP-Elites is capable of evolving the highest performing solutions in addition to generating the largest morphological diversity. Further, MAP-Elites is superior at regaining performance when transferring the population to new and more difficult environments. By analyzing genealogical ancestry we show that MAP-Elites produces more diverse and higher performing stepping stones than the two other objective-based search algorithms. The experiments transitioning the populations to new 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. Together, these results demonstrate the suitability of MAP-elites for the challenging task of morphology-control search for modular robots, and shed light on the algorithm's capability of generating stepping stones for reaching high-performing solutions.

A Transferable Adaptive Domain Adversarial Neural Network for Virtual Reality Augmented EMG-Based Gesture Recognition.

Within the field of electromyography-based (EMG) gesture recognition, disparities exist between the offline accuracy reported in the literature and the real-time usability of a classifier. This gap mainly stems from two factors: 1) The absence of a controller, making the data collected dissimilar to actual control. 2) The difficulty of including the four main dynamic factors (gesture intensity, limb position, electrode shift, and transient changes in the signal), as including their permutations drastically increases the amount of data to be recorded. Contrarily, online datasets are limited to the exact EMG-based controller used to record them, necessitating the recording of a new dataset for each control method or variant to be tested. Consequently, this paper proposes a new type of dataset to serve as an intermediate between offline and online datasets, by recording the data using a real-time experimental protocol. The protocol, performed in virtual reality, includes the four main dynamic factors and uses an EMG-independent controller to guide movements. This EMG-independent feedback ensures that the user is in-the-loop during recording, while enabling the resulting dynamic dataset to be used as an EMG-based benchmark. The dataset is comprised of 20 able-bodied participants completing three to four sessions over a period of 14 to 21 days. The ability of the dynamic dataset to serve as a benchmark is leveraged to evaluate the impact of different recalibration techniques for long-term (across-day) gesture recognition, including a novel algorithm, named TADANN. TADANN consistently and significantly ( [Formula: see text]) outperforms using fine-tuning as the recalibration technique.

Non-flipping DNA glycosylase AlkD scans DNA without formation of a stable interrogation complex.

The multi-step base excision repair (BER) pathway is initiated by a set of enzymes, known as DNA glycosylases, able to scan DNA and detect modified bases among a vast number of normal bases. While DNA glycosylases in the BER pathway generally bend the DNA and flip damaged bases into lesion specific pockets, the HEAT-like repeat DNA glycosylase AlkD detects and excises bases without sequestering the base from the DNA helix. We show by single-molecule tracking experiments that AlkD scans DNA without forming a stable interrogation complex. This contrasts with previously studied repair enzymes that need to flip bases into lesion-recognition pockets and form stable interrogation complexes. Moreover, we show by design of a loss-of-function mutant that the bimodality in scanning observed for the structural homologue AlkF is due to a key structural differentiator between AlkD and AlkF; a positively charged ß-hairpin able to protrude into the major groove of DNA.

A Framework for Automatic Behavior Generation in Multi-Function Swarms.

Multi-function swarms are swarms that solve multiple tasks at once. For example, a quadcopter swarm could be tasked with exploring an area of interest while simultaneously functioning as ad-hoc relays. With this type of multi-function comes the challenge of handling potentially conflicting requirements simultaneously. Using the Quality-Diversity algorithm MAP-elites in combination with a suitable controller structure, a framework for automatic behavior generation in multi-function swarms is proposed. The framework is tested on a scenario with three simultaneous tasks: exploration, communication network creation and geolocation of Radio Frequency (RF) emitters. A repertoire is evolved, consisting of a wide range of controllers, or behavior primitives, with different characteristics and trade-offs in the different tasks. This repertoire enables the swarm to online transition between behaviors featuring different trade-offs of applications depending on the situational requirements. Furthermore, the effect of noise on the behavior characteristics in MAP-elites is investigated. A moderate number of re-evaluations is found to increase the robustness while keeping the computational requirements relatively low. A few selected controllers are examined, and the dynamics of transitioning between these controllers are explored. Finally, the study investigates the importance of individual sensor or controller inputs. This is done through ablation, where individual inputs are disabled and their impact on the performance of the swarm controllers is assessed and analyzed.

Understanding Musical Predictions With an Embodied Interface for Musical Machine Learning.

Machine-learning models of music often exist outside the worlds of musical performance practice and abstracted from the physical gestures of musicians. In this work, we consider how a recurrent neural network (RNN) model of simple music gestures may be integrated into a physical instrument so that predictions are sonically and physically entwined with the performer's actions. We introduce EMPI, an embodied musical prediction interface that simplifies musical interaction and prediction to just one dimension of continuous input and output. The predictive model is a mixture density RNN trained to estimate the performer's next physical input action and the time at which this will occur. Predictions are represented sonically through synthesized audio, and physically with a motorized output indicator. We use EMPI to investigate how performers understand and exploit different predictive models to make music through a controlled study of performances with different models and levels of physical feedback. We show that while performers often favor a model trained on human-sourced data, they find different musical affordances in models trained on synthetic, and even random, data. Physical representation of predictions seemed to affect the length of performances. This work contributes new understandings of how musicians use generative ML models in real-time performance backed up by experimental evidence. We argue that a constrained musical interface can expose the affordances of embodied predictive interactions.

Lamarckian Evolution of Simulated Modular Robots.

We study evolutionary robot systems where not only the robot brains but also the robot bodies are evolvable. Such systems need to include a learning period right after 'birth' to acquire a controller that fits the newly created body. In this paper we investigate the possibility of bootstrapping infant robot learning through employing Lamarckian inheritance of parental controllers. In our system controllers are encoded by a combination of a morphology dependent component, a Central Pattern Generator (CPG), and a morphology independent part, a Compositional Pattern Producing Network (CPPN). This makes it possible to transfer the CPPN part of controllers between different morphologies and to create a Lamarckian system. We conduct experiments with simulated modular robots whose fitness is determined by the speed of locomotion, establish the benefits of inheriting optimized parental controllers, shed light on the conditions that influence these benefits, and observe that changing the way controllers are evolved also impacts the evolved morphologies.

Additive manufacturing of laminar flow cells for single-molecule experiments.

A microfluidic laminar flow cell (LFC) forms an indispensable component in single-molecule experiments, enabling different substances to be delivered directly to the point under observation and thereby tightly controlling the biochemical environment immediately surrounding single molecules. Despite substantial progress in the production of such components, the process remains relatively inefficient, inaccurate and time-consuming. Here we address challenges and limitations in the routines, materials and the designs that have been commonly employed in the field, and introduce a new generation of LFCs designed for single-molecule experiments and assembled using additive manufacturing. We present single- and multi-channel, as well as reservoir-based LFCs produced by 3D printing to perform single-molecule experiments. Using these flow cells along with optical tweezers, we show compatibility with single-molecule experiments including the isolation and manipulation of single DNA molecules either attached to the surface of a coverslip or as freely movable DNA dumbbells, as well as direct observation of protein-DNA interactions. Using additive manufacturing to produce LFCs with versatility of design and ease of production allow experimentalists to optimize the flow cells to their biological experiments and provide considerable potential for performing multi-component single-molecule experiments.

Deep Learning for Electromyographic Hand Gesture Signal Classification Using Transfer Learning.

In recent years, deep learning algorithms have become increasingly more prominent for their unparalleled ability to automatically learn discriminant features from large amounts of data. However, within the field of electromyography-based gesture recognition, deep learning algorithms are seldom employed as they require an unreasonable amount of effort from a single person, to generate tens of thousands of examples. This paper's hypothesis is that general, informative features can be learned from the large amounts of data generated by aggregating the signals of multiple users, thus reducing the recording burden while enhancing gesture recognition. Consequently, this paper proposes applying transfer learning on aggregated data from multiple users while leveraging the capacity of deep learning algorithms to learn discriminant features from large datasets. Two datasets comprised 19 and 17 able-bodied participants, respectively (the first one is employed for pre-training), were recorded for this work, using the Myo armband. A third Myo armband dataset was taken from the NinaPro database and is comprised ten able-bodied participants. Three different deep learning networks employing three different modalities as input (raw EMG, spectrograms, and continuous wavelet transform (CWT)) are tested on the second and third dataset. The proposed transfer learning scheme is shown to systematically and significantly enhance the performance for all three networks on the two datasets, achieving an offline accuracy of 98.31% for 7 gestures over 17 participants for the CWT-based ConvNet and 68.98% for 18 gestures over 10 participants for the raw EMG-based ConvNet. Finally, a use-case study employing eight able-bodied participants suggests that real-time feedback allows users to adapt their muscle activation strategy which reduces the degradation in accuracy normally experienced over time.

Publisher Correction: Breaking the speed limit with multimode fast scanning of DNA by Endonuclease V.

The original version of this Article was updated shortly after publication to add a link to the Peer Review file, which was inadvertently omitted. The Peer Review file is available to download as a Supplementary File from the HTML version of the Article.

Breaking the speed limit with multimode fast scanning of DNA by Endonuclease V.

In order to preserve genomic stability, cells rely on various repair pathways for removing DNA damage. The mechanisms how enzymes scan DNA and recognize their target sites are incompletely understood. Here, by using high-localization precision microscopy along with 133 Hz high sampling rate, we have recorded EndoV and OGG1 interacting with 12-kbp elongated λ-DNA in an optical trap. EndoV switches between three distinct scanning modes, each with a clear range of activation energy barriers. These results concur with average diffusion rate and occupancy of states determined by a hidden Markov model, allowing us to infer that EndoV confinement occurs when the intercalating wedge motif is involved in rigorous probing of the DNA, while highly mobile EndoV may disengage from a strictly 1D helical diffusion mode and hop along the DNA. This makes EndoV the first example of a monomeric, single-conformation and single-binding-site protein demonstrating the ability to switch between three scanning modes.

Evolving locomotion for a 12-DOF quadruped robot in simulated environments.

We demonstrate the power of evolutionary robotics (ER) by comparing to a more traditional approach its performance and cost on the task of simulated robot locomotion. A novel quadruped robot is introduced, the legs of which - each having three non-coplanar degrees of freedom - are very maneuverable. Using a simplistic control architecture and a physics simulation of the robot, gaits are designed both by hand and using a highly parallel evolutionary algorithm (EA). It is found that the EA produces, in a small fraction of the time that takes to design by hand, gaits that travel at two to four times the speed of the hand-designed one. The flexibility of this approach is demonstrated by applying it across a range of differently configured simulators.