Humans seem to have a bias to overshoot when rotating a rotary knob blindfolded around a specified target angle (i.e. during haptic rotation). Whereas some influence factors that strengthen or weaken such an effect are already known, the underlying reasons for the overshoot are still unknown. This work approaches the topic of haptic rotations by analyzing a detailed recording of the movement. We propose an experimental framework and an approach to investigate which upper limb and hand joint movements contribute significantly to a haptic rotation task and to the angle overshoot based on the acquired data. With stepwise regression with backward elimination, we analyze a rotation around 90 degrees counterclockwise with two fingers under different grasping orientations. Our results showed that the wrist joint, the sideways finger movement in the proximal joints, and the distal finger joints contributed significantly to overshooting. This suggests that two phenomena are behind the overshooting: 1) The significant contribution of the wrist joint indicates a bias of a hand-centered egocentric reference frame. 2) Significant contribution of the finger joints indicates a rolling of the fingertips over the rotary knob surface and, thus, a change of contact point for which probably the human does not compensate.

Most commercially available haptic gloves compromise the accuracy of hand-posture measurements in favor of a simpler design with fewer sensors. While inaccurate posture data is often sufficient for the task at hand in biomedical settings such as VR-therapy-aided rehabilitation, measurements should be as precise as possible to digitally recreate hand postures as accurately as possible. With these applications in mind, we have added extra sensors to the commercially available Dexmo haptic glove by Dexta Robotics and applied kinematic models of the haptic glove and the user's hand to improve the accuracy of hand-posture measurements. In this work, we describe the augmentations and the kinematic modeling approach. Additionally, we present and discuss an evaluation of hand posture measurements as a proof of concept.

A typical process of learning to play a piece on a piano consists of a progression through a series of practice units that focus on individual dimensions of the skill, the so-called practice modes. Practice modes in learning to play music comprise a particularly large set of possibilities, such as hand coordination, posture, articulation, ability to read a music score, correct timing or pitch, etc. Self-guided practice is known to be suboptimal, and a model that schedules optimal practice to maximize a learner's progress still does not exist. Because we each learn differently and there are many choices for possible piano practice tasks and methods, the set of practice modes should be dynamically adapted to the human learner, a process typically guided by a teacher. However, having a human teacher guide individual practice is not always feasible since it is time-consuming, expensive, and often unavailable. In this work, we present a modeling framework to guide the human learner through the learning process by choosing the practice modes generated by a policy model. To this end, we present a computational architecture building on a Gaussian process that incorporates 1) the learner state, 2) a policy that selects a suitable practice mode, 3) performance evaluation, and 4) expert knowledge. The proposed policy model is trained to approximate the expert-learner interaction during a practice session. In our future work, we will test different Bayesian optimization techniques, e.g., different acquisition functions, and evaluate their effect on the learning progress.

Diffusion generative models have recently become a robust technique for producing and modifying coherent, high-quality video. This survey offers a systematic overview of critical elements of diffusion models for video generation, covering applications, architectural choices, and the modeling of temporal dynamics. Recent advancements in the field are summarized and grouped into development trends. The survey concludes with an overview of remaining challenges and an outlook on the future of the field.

Physical reasoning is a crucial aspect in the development of general AI systems, given that human learning starts with interacting with the physical world before progressing to more complex concepts. Although researchers have studied and assessed the physical reasoning of AI approaches through various specific benchmarks, there is no comprehensive approach to evaluating and measuring progress. Therefore, we aim to offer an overview of existing benchmarks and their solution approaches and propose a unified perspective for measuring the physical reasoning capacity of AI systems. We select benchmarks that are designed to test algorithmic performance in physical reasoning tasks. While each of the selected benchmarks poses a unique challenge, their ensemble provides a comprehensive proving ground for an AI generalist agent with a measurable skill level for various physical reasoning concepts. This gives an advantage to such an ensemble of benchmarks over other holistic benchmarks that aim to simulate the real world by intertwining its complexity and many concepts. We group the presented set of physical reasoning benchmarks into subcategories so that more narrow generalist AI agents can be tested first on these groups.

The advent of tactile sensors in robotics has sparked many ideas on how robots can leverage direct contact measurements of their environment interactions to improve manipulation tasks. An important line of research in this regard is that of grasp force control, which aims to manipulate objects safely by limiting the amount of force exerted on the object. While prior works have either hand-modeled their force controllers, employed model-based approaches, or have not shown sim-to-real transfer, we propose a model-free deep reinforcement learning approach trained in simulation and then transferred to the robot without further fine-tuning. We therefore present a simulation environment that produces realistic normal forces, which we use to train continuous force control policies. An evaluation in which we compare against a baseline and perform an ablation study shows that our approach outperforms the hand-modeled baseline and that our proposed inductive bias and domain randomization facilitate sim-to-real transfer. Code, models, and supplementary videos are available on https://sites.google.com/view/rl-force-ctrl

Our goal with this survey is to provide an overview of the state of the art deep learning methods for face generation and editing using StyleGAN. The survey covers the evolution of StyleGAN, from PGGAN to StyleGAN3, and explores relevant topics such as suitable metrics for training, different latent representations, GAN inversion to latent spaces of StyleGAN, face image editing, cross-domain face stylization, face restoration, and even Deepfake applications. We aim to provide an entry point into the field for readers that have basic knowledge about the field of deep learning and are looking for an accessible introduction and overview.

In the last few years, artistic image-making with deep learning models has gained a considerable amount of traction. A large number of these models operate directly in the pixel space and generate raster images. This is however not how most humans would produce artworks, for example, by planning a sequence of shapes and strokes to draw. Recent developments in deep learning methods help to bridge the gap between stroke-based paintings and pixel photo generation. With this survey, we aim to provide a structured introduction and understanding of common challenges and approaches in stroke-based rendering algorithms. These algorithms range from simple rule-based heuristics to stroke optimization and deep reinforcement agents, trained to paint images with differentiable vector graphics and neural rendering.

For that, we train an Hourglass network using only feedback from a critic model. The Hourglass network learns to produce a mask to decrease the critic's score of a high score image and increase the critic's score of a low score image by swapping the masked areas between these two images. We trained the model on an imitation learning dataset from the NeurIPS 2020 MineRL Competition Track, where our model learned to mask rewarding objects in a complex interactive 3D environment with a sparse reward signal. This approach was part of the 1st place winning solution in this competition.

Reinforcement learning competitions have formed the basis for standard research benchmarks, galvanized advances in the state-of-the-art, and shaped the direction of the field. Despite this, a majority of challenges suffer from the same fundamental problems: participant solutions to the posed challenge are usually domain-specific, biased to maximally exploit compute resources, and not guaranteed to be reproducible. In this paper, we present a new framework of competition design that promotes the development of algorithms that overcome these barriers. We propose four central mechanisms for achieving this end: submission retraining, domain randomization, desemantization through domain obfuscation, and the limitation of competition compute and environment-sample budget. To demonstrate the efficacy of this design, we proposed, organized, and ran the MineRL 2020 Competition on Sample-Efficient Reinforcement Learning. In this work, we describe the organizational outcomes of the competition and show that the resulting participant submissions are reproducible, non-specific to the competition environment, and sample/resource efficient, despite the difficult competition task.