Search

Soft robots have the potential to revolutionize the use of robotic systems with their capability of establishing safe, robust, and adaptable interactions with their environment, but their precise control remains challenging. In contrast, traditional rigid robots offer high accuracy and repeatability but lack the flexibility of soft robots. We argue that combining these characteristics in a hybrid robotic platform can significantly enhance overall capabilities. This work presents a novel hybrid robotic platform that integrates a rigid manipulator with a fully developed soft arm. This system is equipped with the intelligence necessary to perform flexible and generalizable tasks through imitation learning autonomously. The physical softness and machine learning enable our platform to achieve highly generalizable skills, while the rigid components ensure precision and repeatability., Comment: Corrected missing citation

In this work we study how diffusion-based generative models produce high-dimensional data, such as an image, by implicitly relying on a manifestation of a low-dimensional set of latent abstractions, that guide the generative process. We present a novel theoretical framework that extends NLF, and that offers a unique perspective on SDE-based generative models. The development of our theory relies on a novel formulation of the joint (state and measurement) dynamics, and an information-theoretic measure of the influence of the system state on the measurement process. According to our theory, diffusion models can be cast as a system of SDE, describing a non-linear filter in which the evolution of unobservable latent abstractions steers the dynamics of an observable measurement process (corresponding to the generative pathways). In addition, we present an empirical study to validate our theory and previous empirical results on the emergence of latent abstractions at different stages of the generative process.

Learning from demonstration is a promising way of teaching robots new skills. However, a central problem when executing acquired skills is to recognize risks and failures. This is essential since the demonstrations usually cover only a few mostly successful cases. Inevitable errors during execution require specific reactions that were not apparent in the demonstrations. In this paper, we focus on teaching the robot situational awareness from an initial skill demonstration via kinesthetic teaching and sparse labeling of autonomous skill executions as safe or risky. At runtime, our system, called ILeSiA, detects risks based on the perceived camera images by encoding the images into a low-dimensional latent space representation and training a classifier based on the encoding and the provided labels. In this way, ILeSiA boosts the confidence and safety with which robotic skills can be executed. Our experiments demonstrate that classifiers, trained with only a small amount of user-provided data, can successfully detect numerous risks. The system is flexible because the risk cases are defined by labeling data. This also means that labels can be added as soon as risks are identified by a human supervisor. We provide all code and data required to reproduce our experiments at imitrob.ciirc.cvut.cz/publications/ilesia., Comment: 7 pages, 8 figures

Though there is much interest in fair AI systems, the problem of fairness noncompliance -- which concerns whether fair models are used in practice -- has received lesser attention. Zero-Knowledge Proofs of Fairness (ZKPoF) address fairness noncompliance by allowing a service provider to verify to external parties that their model serves diverse demographics equitably, with guaranteed confidentiality over proprietary model parameters and data. They have great potential for building public trust and effective AI regulation, but no previous techniques for ZKPoF are fit for real-world deployment. We present OATH, the first ZKPoF framework that is (i) deployably efficient with client-facing communication comparable to in-the-clear ML as a Service query answering, and an offline audit phase that verifies an asymptotically constant quantity of answered queries, (ii) deployably flexible with modularity for any score-based classifier given a zero-knowledge proof of correct inference, (iii) deployably secure with an end-to-end security model that guarantees confidentiality and fairness across training, inference, and audits. We show that OATH obtains strong robustness against malicious adversaries at concretely efficient parameter settings. Notably, OATH provides a 1343x improvement to runtime over previous work for neural network ZKPoF, and scales up to much larger models -- even DNNs with tens of millions of parameters.

Diffusion models for Text-to-Image (T2I) conditional generation have seen tremendous success recently. Despite their success, accurately capturing user intentions with these models still requires a laborious trial and error process. This challenge is commonly identified as a model alignment problem, an issue that has attracted considerable attention by the research community. Instead of relying on fine-grained linguistic analyses of prompts, human annotation, or auxiliary vision-language models to steer image generation, in this work we present a novel method that relies on an information-theoretic alignment measure. In a nutshell, our method uses self-supervised fine-tuning and relies on point-wise mutual information between prompts and images to define a synthetic training set to induce model alignment. Our comparative analysis shows that our method is on-par or superior to the state-of-the-art, yet requires nothing but a pre-trained denoising network to estimate MI and a lightweight fine-tuning strategy.

Water's unique anomalies are vital in various applications and biological processes, yet the molecular mechanisms behind these anomalies remain debated, particularly in the metastable liquid phase under supercooling and stretching conditions. Experimental challenges in these conditions have led to simulations suggesting a liquid-liquid phase transition between low-density and high-density water phases, culminating in a liquid-liquid critical point (LLCP). However, these simulations are limited by computational expense, small system sizes, and reliability of water models. Using the FS model, we improve accuracy in predicting water's density and response functions across a broad range of temperatures and pressures. The FS model avoid by design first-order phase transitions towards crystalline phases, allowing thorough exploration of the metastable phase diagram. We employ advanced numerical techniques to bypass dynamical slowing down and perform finite-size scaling on systems significantly larger than those used in previous analyses. Our study extrapolates thermodynamic behavior in the infinite-system limit, accurately demonstrating the existence of the LLCP in the 3D Ising universality class at TC = 186 +/- 4 K and PC = 174 +/- 14 MPa, following a liquid-liquid phase separation below 200 MPa. These predictions align with recent experimental data and more sophisticated models, highlighting that hydrogen bond cooperativity governs the LLCP and the origin of water anomalies. Moreover, we observe that the hydrogen bond network exhibits substantial cooperative fluctuations at scales larger than 10 nm, even at temperatures relevant to biopreservation. These findings have significant implications for fields such as nanotechnology and biophysics, offering new insights into water's behavior under varied conditions., Comment: 18 pages, 3 tables, and 12 figures. Submitted to The Journal of Chemical Physics

Learning from Interactive Demonstrations has revolutionized the way non-expert humans teach robots. It is enough to kinesthetically move the robot around to teach pick-and-place, dressing, or cleaning policies. However, the main challenge is correctly generalizing to novel situations, e.g., different surfaces to clean or different arm postures to dress. This article proposes a novel task parameterization and generalization to transport the original robot policy, i.e., position, velocity, orientation, and stiffness. Unlike the state of the art, only a set of points are tracked during the demonstration and the execution, e.g., a point cloud of the surface to clean. We then propose to fit a non-linear transformation that would deform the space and then the original policy using the paired source and target point sets. The use of function approximators like Gaussian Processes allows us to generalize, or transport, the policy from every space location while estimating the uncertainty of the resulting policy due to the limited points in the task parameterization point set and the reduced number of demonstrations. We compare the algorithm's performance with state-of-the-art task parameterization alternatives and analyze the effect of different function approximators. We also validated the algorithm on robot manipulation tasks, i.e., different posture arm dressing, different location product reshelving, and different shape surface cleaning.

In the upcoming 6G era, mobile networks must deal with more challenging applications (e.g., holographic telepresence and immersive communication) and meet far more stringent application requirements stemming along the edge-cloud continuum. These new applications will create an elevated level of expectations on performance, reliability, ubiquity, trustworthiness, security, openness, and sustainability, pushing the boundaries of innovation and driving transformational change across the architecture of future mobile networks. Towards this end, ADROIT6G proposes a set of disruptive innovations with a clear vision on setting a 6G network architecture that can be tailored to the requirements of innovative applications and match the ambitious KPIs set for 6G networks. More specifically, the key transformations that ADROIT6G considers essential to 6G network evolution are: i) AI/ML-powered optimisations across the network, exploring solutions in the "Distributed Artificial Intelligence (DAI)" domain for high performance and automation; ii) Transforming to fully cloud-native network software, which can be implemented across various edge-cloud platforms, with security built integrally into the network user plan; and iii) Software driven, zero-touch operations and ultimately automation of every aspect of the network and the services it delivers.

The analysis of scientific data and complex multivariate systems requires information quantities that capture relationships among multiple random variables. Recently, new information-theoretic measures have been developed to overcome the shortcomings of classical ones, such as mutual information, that are restricted to considering pairwise interactions. Among them, the concept of information synergy and redundancy is crucial for understanding the high-order dependencies between variables. One of the most prominent and versatile measures based on this concept is O-information, which provides a clear and scalable way to quantify the synergy-redundancy balance in multivariate systems. However, its practical application is limited to simplified cases. In this work, we introduce S$\Omega$I, which allows for the first time to compute O-information without restrictive assumptions about the system. Our experiments validate our approach on synthetic data, and demonstrate the effectiveness of S$\Omega$I in the context of a real-world use case.

Collaborative machine learning (ML) is widely used to enable institutions to learn better models from distributed data. While collaborative approaches to learning intuitively protect user data, they remain vulnerable to either the server, the clients, or both, deviating from the protocol. Indeed, because the protocol is asymmetric, a malicious server can abuse its power to reconstruct client data points. Conversely, malicious clients can corrupt learning with malicious updates. Thus, both clients and servers require a guarantee when the other cannot be trusted to fully cooperate. In this work, we propose a peer-to-peer (P2P) learning scheme that is secure against malicious servers and robust to malicious clients. Our core contribution is a generic framework that transforms any (compatible) algorithm for robust aggregation of model updates to the setting where servers and clients can act maliciously. Finally, we demonstrate the computational efficiency of our approach even with 1-million parameter models trained by 100s of peers on standard datasets., Comment: Accepted at NeurIPS 2023

In this work we present a new method for the estimation of Mutual Information (MI) between random variables. Our approach is based on an original interpretation of the Girsanov theorem, which allows us to use score-based diffusion models to estimate the Kullback Leibler divergence between two densities as a difference between their score functions. As a by-product, our method also enables the estimation of the entropy of random variables. Armed with such building blocks, we present a general recipe to measure MI, which unfolds in two directions: one uses conditional diffusion process, whereas the other uses joint diffusion processes that allow simultaneous modelling of two random variables. Our results, which derive from a thorough experimental protocol over all the variants of our approach, indicate that our method is more accurate than the main alternatives from the literature, especially for challenging distributions. Furthermore, our methods pass MI self-consistency tests, including data processing and additivity under independence, which instead are a pain-point of existing methods.

Lipid nanoparticles own a remarkable potential in nanomedicine, only partially disclosed. While the clinical use of liposomes and cationic lipid-nucleic acid complexes is well-established, liquid lipid nanoparticles (nanoemulsions), solid lipid nanoparticles, and nanostructured lipid carriers have even greater potential. However, they face obstacles in being used in clinics due to a lack of understanding about the molecular mechanisms controlling their drug loading and release, interactions with the biological environment (such as the protein corona), and shelf-life stability. To create effective drug delivery carriers and successfully translate bench research to clinical settings, it is crucial to have a thorough understanding of the internal structure of lipid nanoparticles. Through synchrotron small-angle X-ray scattering experiments, we determined the spatial distribution and internal structure of the nanoparticles' lipid, surfactant, and the water in them. The nanoparticles themselves have a barrel-like shape that consists of coplanar lipid platelets (specifically cetyl palmitate) that are partially covered by polysorbate 80 surfactant and retain a small amount of hydration water. Although the platelet structure was expected, the presence of surfactant molecules forming sticky patches between adjacent platelets challenges the classical core-shell model used to describe solid lipid nanoparticles. Additionally, the surfactant partially covers the water-nanoparticle interface, allowing certain lipid regions to come into direct contact with surrounding water. These structural features play a significant role in drug loading and release, biological fluid interaction, and nanoparticle stability, making these findings valuable for the rational design of lipid-based nanoparticles., Comment: 22 pages, 11 figures

To control how a robot moves, motion planning algorithms must compute paths in high-dimensional state spaces while accounting for physical constraints related to motors and joints, generating smooth and stable motions, avoiding obstacles, and preventing collisions. A motion planning algorithm must therefore balance competing demands, and should ideally incorporate uncertainty to handle noise, model errors, and facilitate deployment in complex environments. To address these issues, we introduce a framework for robot motion planning based on variational Gaussian processes, which unifies and generalizes various probabilistic-inference-based motion planning algorithms, and connects them with optimization-based planners. Our framework provides a principled and flexible way to incorporate equality-based, inequality-based, and soft motion-planning constraints during end-to-end training, is straightforward to implement, and provides both interval-based and Monte-Carlo-based uncertainty estimates. We conduct experiments using different environments and robots, comparing against baseline approaches based on the feasibility of the planned paths, and obstacle avoidance quality. Results show that our proposed approach yields a good balance between success rates and path quality., Comment: Code and supplementary video available at: https://github.com/luke-ck/vgpmp

Efficient and sustainable techniques for separating water-methanol mixtures are in high demand in the industry. Recent studies have revealed that membranes and 2D materials could achieve such separation. In our research, we explore the impact of a nanoconfining graphene slit-pore on the dynamics and structure of water-methanol mixtures. By Molecular Dynamics simulations of a coarse-grained model for water mixtures containing up to 25% methanol, we show that, for appropriate pore sizes, water tends to occupy the center of the pore. In contrast, methanol's apolar moiety accumulates near the hydrophobic walls. Additionally, modifying the pore's width leads to a non-monotonic change in the diffusivity of each component. However, water always diffuses faster than methanol, implying that it should be possible to identify an optimal configuration for water-methanol separation based on physical mechanisms. Our calculations indicate that one of the more effective pore sizes, 12.5{\AA}, is also mechanically stable, minimizing the energy cost of a possible filtering membrane.

Multi-modal data-sets are ubiquitous in modern applications, and multi-modal Variational Autoencoders are a popular family of models that aim to learn a joint representation of the different modalities. However, existing approaches suffer from a coherence-quality tradeoff, where models with good generation quality lack generative coherence across modalities, and vice versa. We discuss the limitations underlying the unsatisfactory performance of existing methods, to motivate the need for a different approach. We propose a novel method that uses a set of independently trained, uni-modal, deterministic autoencoders. Individual latent variables are concatenated into a common latent space, which is fed to a masked diffusion model to enable generative modeling. We also introduce a new multi-time training method to learn the conditional score network for multi-modal diffusion. Our methodology substantially outperforms competitors in both generation quality and coherence, as shown through an extensive experimental campaign.

Generative Models (GMs) have attracted considerable attention due to their tremendous success in various domains, such as computer vision where they are capable to generate impressive realistic-looking images. Likelihood-based GMs are attractive due to the possibility to generate new data by a single model evaluation. However, they typically achieve lower sample quality compared to state-of-the-art score-based diffusion models (DMs). This paper provides a significant step in the direction of addressing this limitation. The idea is to borrow one of the strengths of score-based DMs, which is the ability to perform accurate density estimation in low-density regions and to address manifold overfitting by means of data mollification. We connect data mollification through the addition of Gaussian noise to Gaussian homotopy, which is a well-known technique to improve optimization. Data mollification can be implemented by adding one line of code in the optimization loop, and we demonstrate that this provides a boost in generation quality of likelihood-based GMs, without computational overheads. We report results on image data sets with popular likelihood-based GMs, including variants of variational autoencoders and normalizing flows, showing large improvements in FID score., Comment: NeurIPS 2023

Radiotherapy (RT) is a key component in the treatment of various cancers, including Acute Lymphocytic Leukemia (ALL) and Acute Myelogenous Leukemia (AML). Precise delineation of organs at risk (OARs) and target areas is essential for effective treatment planning. Intensity Modulated Radiotherapy (IMRT) techniques, such as Total Marrow Irradiation (TMI) and Total Marrow and Lymph node Irradiation (TMLI), provide more precise radiation delivery compared to Total Body Irradiation (TBI). However, these techniques require time-consuming manual segmentation of structures in Computerized Tomography (CT) scans by the Radiation Oncologist (RO). In this paper, we present a deep learning-based auto-contouring method for segmenting Planning Target Volume (PTV) for TMLI treatment using the U-Net architecture. We trained and compared two segmentation models with two different loss functions on a dataset of 100 patients treated with TMLI at the Humanitas Research Hospital between 2011 and 2021. Despite challenges in lymph node areas, the best model achieved an average Dice score of 0.816 for PTV segmentation. Our findings are a preliminary but significant step towards developing a segmentation model that has the potential to save radiation oncologists a considerable amount of time. This could allow for the treatment of more patients, resulting in improved clinical practice efficiency and more reproducible contours.

In the medical images field, semantic segmentation is one of the most important, yet difficult and time-consuming tasks to be performed by physicians. Thanks to the recent advancement in the Deep Learning models regarding Computer Vision, the promise to automate this kind of task is getting more and more realistic. However, many problems are still to be solved, like the scarce availability of data and the difficulty to extend the efficiency of highly specialised models to general scenarios. Organs at risk segmentation for radiotherapy treatment planning falls in this category, as the limited data available negatively affects the possibility to develop general-purpose models; in this work, we focus on the possibility to solve this problem by presenting three types of ensembles of single-organ models able to produce multi-organ masks exploiting the different specialisations of their components. The results obtained are promising and prove that this is a possible solution to finding efficient multi-organ segmentation methods.

The separation of liquid mixture components is relevant to many applications -- ranging from water purification to biofuel production -- and is a growing concern related to the UN Sustainable Development Goals (SDGs), such as ``Clean water and Sanitation'' and ``Affordable and clean energy''. One promising technique is using graphene slit-pores as filters, or sponges, because the confinement potentially affects the properties of the mixture components in different ways, favoring their separation. However, no systematic study has shown how the size of a pore changes the thermodynamics of the surrounding mixture. Here, we focus on water-methanol mixtures and explore, using Molecular Dynamics simulations, the effects of a graphene pore, with size ranging from 6.5 to 13 \AA, for three compositions: pure water, 90\%-10\%, and 75\%-25\% water-methanol. We show that tuning the pore size can change the mixture pressure, density and composition in bulk due to the size-dependent methanol sequestration within the pore. Our results can help in optimizing the graphene pore size for filtering applications.

We introduce Functional Diffusion Processes (FDPs), which generalize score-based diffusion models to infinite-dimensional function spaces. FDPs require a new mathematical framework to describe the forward and backward dynamics, and several extensions to derive practical training objectives. These include infinite-dimensional versions of Girsanov theorem, in order to be able to compute an ELBO, and of the sampling theorem, in order to guarantee that functional evaluations in a countable set of points are equivalent to infinite-dimensional functions. We use FDPs to build a new breed of generative models in function spaces, which do not require specialized network architectures, and that can work with any kind of continuous data. Our results on real data show that FDPs achieve high-quality image generation, using a simple MLP architecture with orders of magnitude fewer parameters than existing diffusion models., Comment: Under review

Unstructured proteins can modulate cellular responses to environmental conditions by undergoing coil-globule transitions and phase separation. However, the molecular mechanisms of these phenomena still need to be fully understood. Here, we use Monte Carlo calculations of a coarse-grained model incorporating water's effects on the system's free energy. Following previous studies, we model an unstructured protein as a polymer chain. Because we are interested in investigating how it responds to thermodynamic changes near a hydrophobic surface under different conditions, we choose an entirely hydrophobic sequence to maximize the interaction with the interface. We show that a slit pore confinement without top-down symmetry enhances the unfolding and adsorption of the chain in both random coil and globular states. Moreover, we demonstrate that the hydration water modulates this behavior depending on the thermodynamic parameters. Our findings provide insights into how homopolymers, and possibly unstructured proteins, can sense and adjust to external stimuli such as nanointerfaces or stresses., Comment: 25 pages, 5 figures,8 supplementary movies (mp4)

Developing physically assistive robots capable of dressing assistance has the potential to significantly improve the lives of the elderly and disabled population. However, most robotics dressing strategies considered a single robot only, which greatly limited the performance of the dressing assistance. In fact, healthcare professionals perform the task bimanually. Inspired by them, we propose a bimanual cooperative scheme for robotic dressing assistance. In the scheme, an interactive robot joins hands with the human thus supporting/guiding the human in the dressing process, while the dressing robot performs the dressing task. We identify a key feature that affects the dressing action and propose an optimal strategy for the interactive robot using the feature. A dressing coordinate based on the posture of the arm is defined to better encode the dressing policy. We validate the interactive dressing scheme with extensive experiments and also an ablation study. The experiment video is available on https://sites.google.com/view/bimanualassitdressing/home

Interactive Imitation Learning (IIL) is a branch of Imitation Learning (IL) where human feedback is provided intermittently during robot execution allowing an online improvement of the robot's behavior. In recent years, IIL has increasingly started to carve out its own space as a promising data-driven alternative for solving complex robotic tasks. The advantages of IIL are its data-efficient, as the human feedback guides the robot directly towards an improved behavior, and its robustness, as the distribution mismatch between the teacher and learner trajectories is minimized by providing feedback directly over the learner's trajectories. Nevertheless, despite the opportunities that IIL presents, its terminology, structure, and applicability are not clear nor unified in the literature, slowing down its development and, therefore, the research of innovative formulations and discoveries. In this article, we attempt to facilitate research in IIL and lower entry barriers for new practitioners by providing a survey of the field that unifies and structures it. In addition, we aim to raise awareness of its potential, what has been accomplished and what are still open research questions. We organize the most relevant works in IIL in terms of human-robot interaction (i.e., types of feedback), interfaces (i.e., means of providing feedback), learning (i.e., models learned from feedback and function approximators), user experience (i.e., human perception about the learning process), applications, and benchmarks. Furthermore, we analyze similarities and differences between IIL and RL, providing a discussion on how the concepts offline, online, off-policy and on-policy learning should be transferred to IIL from the RL literature. We particularly focus on robotic applications in the real world and discuss their implications, limitations, and promising future areas of research.

Performing bimanual tasks with dual robotic setups can drastically increase the impact on industrial and daily life applications. However, performing a bimanual task brings many challenges, like synchronization and coordination of the single-arm policies. This article proposes the Safe, Interactive Movement Primitives Learning (SIMPLe) algorithm, to teach and correct single or dual arm impedance policies directly from human kinesthetic demonstrations. Moreover, it proposes a novel graph encoding of the policy based on Gaussian Process Regression (GPR) where the single-arm motion is guaranteed to converge close to the trajectory and then towards the demonstrated goal. Regulation of the robot stiffness according to the epistemic uncertainty of the policy allows for easily reshaping the motion with human feedback and/or adapting to external perturbations. We tested the SIMPLe algorithm on a real dual-arm setup where the teacher gave separate single-arm demonstrations and then successfully synchronized them only using kinesthetic feedback or where the original bimanual demonstration was locally reshaped to pick a box at a different height.

Many high precision (dis)assembly tasks are still being performed by humans, whereas this is an ideal opportunity for automation. This paper provides a framework which enables a non-expert human operator to teach a robotic arm to do complex precision tasks. The framework uses a variable Cartesian impedance controller to execute trajectories learned from kinesthetic human demonstrations. Feedback can be given to interactively reshape or speed up the original demonstration. Board localization is done through a visual estimation of the task board position and refined through haptic feedback. Our framework is tested on the Robothon benchmark disassembly challenge, where the robot has to perform complex precision tasks, such as a key insertion. The results show high success rates for each of the manipulation subtasks, including cases when the box is in novel poses. An ablation study is also performed to evaluate the components of the framework., Comment: Under review for ICRA 2023