Concept-based explainable artificial intelligence (C-XAI) can let people see which representations an AI model has learned. This is particularly important when high-level semantic information (e.g., actions and relations) is used to make decisions about abstract categories (e.g., danger). In such tasks, AI models need to generalise beyond situation-specific details, and this ability can be reflected in C-XAI outputs that randomise over irrelevant features. However, it is unclear whether people appreciate such generalisation and can distinguish it from other, less desirable forms of imprecision in C-XAI outputs. Therefore, the present study investigated how the generality and relevance of C-XAI outputs affect people's evaluation of AI. In an experimental railway safety evaluation scenario, participants rated the performance of a simulated AI that classified traffic scenes involving people as dangerous or not. These classification decisions were explained via concepts in the form of similar image snippets. The latter differed in their match with the classified image, either regarding a highly relevant feature (i.e., people's relation to tracks) or a less relevant feature (i.e., people's action). Contrary to the hypotheses, concepts that generalised over less relevant features were rated lower than concepts that matched the classified image precisely. Moreover, their ratings were no better than those for systematic misrepresentations of the less relevant feature. Conversely, participants were highly sensitive to imprecisions in relevant features. These findings cast doubts on the assumption that people can easily infer from C-XAI outputs whether AI models have gained a deeper understanding of complex situations.

Protein design and engineering has become a crucial component of pharmaceutical research and development, finding application in a wide variety of diagnostic and industrial settings.

Inverse Protein Folding (IPF) is a critical subtask in the field of protein design, aiming to engineer amino acid sequences capable of folding correctly into a specified three-dimensional (3D) conformation. Although substantial progress has been achieved in recent years, existing methods generally rely on either backbone coordinates or molecular surface features alone, which restricts their ability to fully capture the complex chemical and geometric constraints necessary for precise sequence prediction. To address this limitation, we present DS-ProGen, a dual-structure deep language model for functional protein design, which integrates both backbone geometry and surface-level representations. By incorporating backbone coordinates as well as surface chemical and geometric descriptors into a next-amino-acid prediction paradigm, DS-ProGen is able to generate functionally relevant and structurally stable sequences while satisfying both global and local conformational constraints. On the PRIDE dataset, DS-ProGen attains the current state-of-the-art recovery rate of 61.47%, demonstrating the synergistic advantage of multi-modal structural encoding in protein design. Furthermore, DS-ProGen excels in predicting interactions with a variety of biological partners, including ligands, ions, and RNA, confirming its robust functional retention capabilities.

One of the LPDI's de novo protein binders (red) bound to the protein Bcl2 (blue) in complex with FDA-approved drug Venetoclax (beige) LPDI EPFL In 2023, scientists in the joint School of Engineering and School of Life Sciences Laboratory of Protein Design and Immunoengineering (LPDI), led by Bruno Correia, published a deep-learning pipeline for designing new proteins to interact with therapeutic targets. MaSIF can rapidly scan millions of proteins to identify optimal matches between molecules based on their chemical and geometric surface properties, enabling scientists to engineer novel protein-protein interactions that play key roles in cell regulation and therapeutics. A year and a half later, the team has reported an exciting advancement of this technology. They have used MaSIF to design novel protein binders to interact with known protein complexes involving small molecules like therapeutic drugs or hormones. Because these bound small molecules induce subtle changes in the surface properties ('neosurfaces') of these protein-drug complexes, they can act as'on' or'off' switches for the fine control of cellular functions like DNA transcription or protein degradation.

Accurate grasping is the key to several robotic tasks including assembly and household robotics. Executing a successful grasp in a cluttered environment requires multiple levels of scene understanding: First, the robot needs to analyze the geometric properties of individual objects to find feasible grasps. These grasps need to be compliant with the local object geometry. Second, for each proposed grasp, the robot needs to reason about the interactions with other objects in the scene. Finally, the robot must compute a collision-free grasp trajectory while taking into account the geometry of the target object. Most grasp detection algorithms directly predict grasp poses in a monolithic fashion, which does not capture the composability of the environment. In this paper, we introduce an end-to-end architecture for object-centric grasping. The method uses pointcloud data from a single arbitrary viewing direction as an input and generates an instance-centric representation for each partially observed object in the scene. This representation is further used for object reconstruction and grasp detection in cluttered table-top scenes. We show the effectiveness of the proposed method by extensively evaluating it against state-of-the-art methods on synthetic datasets, indicating superior performance for grasping and reconstruction. Additionally, we demonstrate real-world applicability by decluttering scenes with varying numbers of objects.

In unstructured outdoor environments, robotics requires accurate and efficient odometry with low computational time. Existing low-bias LiDAR odometry methods are often computationally expensive. To address this problem, we present a lightweight LiDAR odometry method that converts unorganized point cloud data into a spherical range image (SRI) and filters out surface, edge, and ground features in the image plane. This substantially reduces computation time and the required features for odometry estimation in LOAM-based algorithms. Our odometry estimation method does not rely on global maps or loop closure algorithms, which further reduces computational costs. Experimental results generate a translation and rotation error of 0.86\% and 0.0036{\deg}/m on the KITTI dataset with an average runtime of 78ms. In addition, we tested the method with our data, obtaining an average closed-loop error of 0.8m and a runtime of 27ms over eight loops covering 3.5Km.

Robotic dexterous grasping is a challenging problem due to the high degree of freedom (DoF) and complex contacts of multi-fingered robotic hands. Existing deep reinforcement learning (DRL) based methods leverage human demonstrations to reduce sample complexity due to the high dimensional action space with dexterous grasping. However, less attention has been paid to hand-object interaction representations for high-level generalization. In this paper, we propose a novel geometric and spatial hand-object interaction representation, named DexRep, to capture dynamic object shape features and the spatial relations between hands and objects during grasping. DexRep comprises Occupancy Feature for rough shapes within sensing range by moving hands, Surface Feature for changing hand-object surface distances, and Local-Geo Feature for local geometric surface features most related to potential contacts. Based on the new representation, we propose a dexterous deep reinforcement learning method to learn a generalizable grasping policy DexRepNet. Experimental results show that our method outperforms baselines using existing representations for robotic grasping dramatically both in grasp success rate and convergence speed. It achieves a 93% grasping success rate on seen objects and higher than 80% grasping success rates on diverse objects of unseen categories in both simulation and real-world experiments.

The search for ephemeral liquid water on Mars is an ongoing activity. After the recession of the seasonal polar ice cap on Mars, small water ice patches may be left behind in shady places due to the low thermal conductivity of the Martian surface and atmosphere. During late spring and early summer, these patches may be exposed to direct sunlight and warm up rapidly enough for the liquid phase to emerge. To see the spatial and temporal occurrence of such ice patches, optical images should be searched for and checked. Previously a manual image analysis was conducted on 110 images from the southern hemisphere, captured by the High Resolution Imaging Science Experiment (HiRISE) camera onboard the Mars Reconnaissance Orbiter space mission. Out of these, 37 images were identified with smaller ice patches, which were distinguishable by their brightness, colour and strong connection to local topographic shading. In this study, a convolutional neural network (CNN) is applied to find further images with potential water ice patches in the latitude band between -40{\deg} and -60{\deg}, where the seasonal retreat of the polar ice cap happens. Previously analysed HiRISE images were used to train the model, where each image was split into hundreds of pieces (chunks), expanding the training dataset to 6240 images. A test run conducted on 38 new HiRISE images indicates that the program can generally recognise small bright patches, however further training might be needed for more precise identification. This further training has been conducted now, incorporating the results of the previous test run. To retrain the model, 18646 chunks were analysed and 48 additional epochs were ran. In the end the model produced a 94% accuracy in recognising ice, 58% of these images showed small enough ice patches on them. The rest of the images was covered by too much ice or showed CO2 ice sublimation in some places.

An event-based camera outputs an event whenever a change in scene brightness of a preset magnitude is detected at a particular pixel location in the sensor plane. The resulting sparse and asynchronous output, coupled with the high dynamic range and temporal resolution of this novel camera, motivate the study of event-based cameras for navigation and landing applications. However, the lack of real-world and synthetic datasets to support this line of research has limited its consideration for onboard use. This paper presents a methodology and a software pipeline for generating event-based vision datasets from optimal landing trajectories during the approach of a target body. We construct sequences of photorealistic images of the lunar surface with the Planet and Asteroid Natural Scene Generation Utility (PANGU) at different viewpoints along a set of optimal descent trajectories obtained by varying the boundary conditions. The generated image sequences are then converted into event streams by means of an event-based camera emulator. We demonstrate that the pipeline can generate realistic event-based representations of surface features by constructing a dataset of 500 trajectories, complete with event streams and motion field ground truth data. We anticipate that novel event-based vision datasets can be generated using this pipeline to support various spacecraft pose reconstruction problems given events as input, and we hope that the proposed methodology would attract the attention of researchers working at the intersection of neuromorphic vision and guidance navigation and control.