Related papers: Physical Neural Cellular Automata for 2D Shape Cla…
Cellular automata have long been celebrated for their ability to generate complex behaviors from simple, local rules, with well-known discrete models like Conway's Game of Life proven capable of universal computation. Recent advancements…
Neural networks leverage robust internal representations in order to generalise. Learning them is difficult, and often requires a large training set that covers the data distribution densely. We study a common setting where our task is not…
The ongoing deep learning revolution has allowed computers to outclass humans in various games and perceive features imperceptible to humans during classification tasks. Current machine learning techniques have clearly distinguished…
Template 3D shapes are useful for many tasks in graphics and vision, including fitting observation data, analyzing shape collections, and transferring shape attributes. Because of the variety of geometry and topology of real-world shapes,…
Image segmentation is widely used in a variety of computer vision tasks, such as object localization and recognition, boundary detection, and medical imaging. This thesis proposes deep learning architectures to improve automatic object…
Legged robots have significant potential to operate in highly unstructured environments. The design of locomotion control is, however, still challenging. Currently, controllers must be either manually designed for specific robots and tasks,…
While today's robots are able to perform sophisticated tasks, they can only act on objects they have been trained to recognize. This is a severe limitation: any robot will inevitably see new objects in unconstrained settings, and thus will…
Machine learning has made tremendous progress in recent years and received large amounts of public attention. Though we are still far from designing a full artificially intelligent agent, machine learning has brought us many applications in…
To learn object models for robotic manipulation, unsupervised methods cannot provide accurate object structural information and supervised methods require a large amount of manually labeled training samples, thus interactive object…
Many soft-body organisms found in nature flourish underwater. Similarly, soft robots are potentially well-suited for underwater environments partly because the problematic effects of gravity, friction, and harmonic oscillations are less…
The goal of machine learning is to provide solutions which are trained by data or by experience coming from the environment. Many training algorithms exist and some brilliant successes were achieved. But even in structured environments for…
Many computer models such as cellular automata and artificial neural networks have been developed and successfully applied. However, in some cases, these models might be restrictive on the possible solutions or their solutions might be…
Tools, models and statistical methods for signal processing and medical image analysis and training deep learning models to create research prototypes for eventual clinical applications are of special interest to the biomedical imaging…
The problem of generating microstructures of complex materials in silico has been approached from various directions including simulation, Markov, deep learning and descriptor-based approaches. This work presents a hybrid method that is…
Machine learning approaches to spatiotemporal physical systems have primarily focused on next-frame prediction, with the goal of learning an accurate emulator for the system's evolution in time. However, these emulators are computationally…
Data generated by edge devices has the potential to train intelligent autonomous systems across various domains. Despite the emergence of diverse machine learning approaches addressing privacy concerns and utilizing distributed data,…
Neural Cellular Automata (NCAs) have been proven effective in simulating morphogenetic processes, the continuous construction of complex structures from very few starting cells. Recent developments in NCAs lie in the 2D domain, namely…
Discovering and optimizing commercially viable materials for clean energy applications typically takes over a decade. Self-driving laboratories that iteratively design, execute, and learn from material science experiments in a fully…
We show that neural networks trained by evolutionary reinforcement learning can enact efficient molecular self-assembly protocols. Presented with molecular simulation trajectories, networks learn to change temperature and chemical potential…
We propose the concept of autonomous self-damaging in "smart" composite materials, controlled by activation of added nanosize "damaging" capsules. Percolation-type modeling approach earlier applied to the related concept of self-healing…