Robot design optimization, imitation learning and system identification share a common problem which requires optimization over robot or task parameters at the same time as optimizing the robot motion. To solve these problems, we can use differentiable optimal control for which the gradients of the robot's motion with respect to the parameters are required. We propose a method to efficiently compute these gradients analytically via the differential dynamic programming (DDP) algorithm using sensitivity analysis (SA). We show that we must include second-order dynamics terms when computing the gradients. However, we do not need to include them when computing the motion. We validate our approach on the pendulum and double pendulum systems. Furthermore, we compare against using the derivatives of the iterative linear quadratic regulator (iLQR), which ignores these second-order terms everywhere, on a co-design task for the Kinova arm, where we optimize the link lengths of the robot for a target reaching task. We show that optimizing using iLQR gradients diverges as ignoring the second-order dynamics affects the computation of the derivatives. Instead, optimizing using DDP gradients converges to the same optimum for a range of initial designs allowing our formulation to scale to complex systems.

Graph Neural Networks achieve state-of-the-art performance on a plethora of graph classification tasks, especially due to pooling operators, which aggregate learned node embeddings hierarchically into a final graph representation. However, they are not only questioned by recent work showing on par performance with random pooling, but also ignore completely higher-order connectivity patterns. To tackle this issue, we propose HoscPool, a clustering-based graph pooling operator that captures higher-order information hierarchically, leading to richer graph representations. In fact, we learn a probabilistic cluster assignment matrix end-to-end by minimising relaxed formulations of motif spectral clustering in our objective function, and we then extend it to a pooling operator. We evaluate HoscPool on graph classification tasks and its clustering component on graphs with ground-truth community structure, achieving best performance. Lastly, we provide a deep empirical analysis of pooling operators' inner functioning.

In this work, we explore recurrent neural network architectures for tuberculosis (TB) cough classification. In contrast to previous unsuccessful attempts to implement deep architectures in this domain, we show that a basic bidirectional long short-term memory network (BiLSTM) can achieve improved performance. In addition, we show that by performing greedy feature selection in conjunction with a newly-proposed attention-based architecture that learns patient invariant features, substantially better generalisation can be achieved compared to a baseline and other considered architectures. Furthermore, this attention mechanism allows an inspection of the temporal regions of the audio signal considered to be important for classification to be performed. Finally, we develop a neural style transfer technique to infer idealised inputs which can subsequently be analysed. We find distinct differences between the idealised power spectra of TB and non-TB coughs, which provide clues about the origin of the features in the audio signal.

An ensemble of classifiers combines several single classifiers to deliver a final prediction or classification decision. An increasingly provoking question is whether such systems can outperform the single best classifier. If so, what form of an ensemble of classifiers (also known as multiple classifier learning systems or multiple classifiers) yields the most significant benefits in the size or diversity of the ensemble itself? Given that the tests used to detect autism traits are time-consuming and costly, developing a system that will provide the best outcome and measurement of autism spectrum disorder (ASD) has never been critical. In this paper, several single and later multiple classifiers learning systems are evaluated in terms of their ability to predict and identify factors that influence or contribute to ASD for early screening purposes. A dataset of behavioural data and robot-enhanced therapy of 3,000 sessions and 300 hours, recorded from 61 children are utilised for this task. Simulation results show the superior predictive performance of multiple classifier learning systems (especially those with three classifiers per ensemble) compared to individual classifiers, with bagging and boosting achieving excellent results. It also appears that social communication gestures remain the critical contributing factor to the ASD problem among children.

We present a technique for zero-shot generation of a 3D model using only a target text prompt. Without any 3D supervision our method deforms the control shape of a limit subdivided surface along with its texture map and normal map to obtain a 3D asset that corresponds to the input text prompt and can be easily deployed into games or modeling applications. We rely only on a pre-trained CLIP model that compares the input text prompt with differentiably rendered images of our 3D model. While previous works have focused on stylization or required training of generative models we perform optimization on mesh parameters directly to generate shape, texture or both. To constrain the optimization to produce plausible meshes and textures we introduce a number of techniques using image augmentations and the use of a pretrained prior that generates CLIP image embeddings given a text embedding.

After plenty of leaks, Ubisoft has confirmed that Assassin's Creed Mirage is the next entry in its long-running series. More details are expected to drop during the Ubisoft Forward event September 10th, but for now we can gleam some tidbits from the announcement image. It shows Basim Ibn Ishaq, a character from the recent Assassin's Creed Valhalla, leaping with his hidden blade in front of the Palace of the Golden Gate in Baghdad (via Polygon). That lines up with previous leaks around the game's setting, which also indicated that Mirage would be a return to stealth gameplay for the series. The new title was originally intended to be DLC for Valhalla, but Bloomberg reports that it was later transformed into a standalone experience to fill out Ubisoft's release schedule. No matter its conception, it's nice to see the series return to its Middle Eastern roots.

Were you unable to attend Transform 2022? Check out all of the summit sessions in our on-demand library now! A few years ago, a New Jersey man was arrested for shoplifting and spent ten days in jail. He was actually 30 miles away during the time of the incident; police facial recognition software wrongfully identified him. Facial recognition's race and gender failings are well known.

Inspired by human behavior when traveling over unknown terrain, this study proposes the use of probing strategies and integrates them into a traversability analysis framework to address safe navigation on unknown rough terrain. Our framework integrates collapsibility information into our existing traversability analysis, as vision and geometric information alone could be misled by unpredictable non-rigid terrains such as soft soil, bush area, or water puddles. With the new traversability analysis framework, our robot has a more comprehensive assessment of unpredictable terrain, which is critical for its safety in outdoor environments. The pipeline first identifies the terrain's geometric and semantic properties using an RGB-D camera and desired probing locations on questionable terrains. These regions are probed using a force sensor to determine the risk of terrain collapsing when the robot steps over it. This risk is formulated as a collapsibility metric, which estimates an unpredictable region's ground collapsibility. Thereafter, the collapsibility metric, together with geometric and semantic spatial data, is combined and analyzed to produce global and local traversability grid maps. These traversability grid maps tell the robot whether it is safe to step over different regions of the map. The grid maps are then utilized to generate optimal paths for the robot to safely navigate to its goal. Our approach has been successfully verified on a quadrupedal robot in both simulation and real-world experiments.

Computational phenotyping allows for unsupervised discovery of subgroups of patients as well as corresponding co-occurring medical conditions from electronic health records (EHR). Typically, EHR data contains demographic information, diagnoses and laboratory results. Discovering (novel) phenotypes has the potential to be of prognostic and therapeutic value. Providing medical practitioners with transparent and interpretable results is an important requirement and an essential part for advancing precision medicine. Low-rank data approximation methods such as matrix (e.g., non-negative matrix factorization) and tensor decompositions (e.g., CANDECOMP/PARAFAC) have demonstrated that they can provide such transparent and interpretable insights. Recent developments have adapted low-rank data approximation methods by incorporating different constraints and regularizations that facilitate interpretability further. In addition, they offer solutions for common challenges within EHR data such as high dimensionality, data sparsity and incompleteness. Especially extracting temporal phenotypes from longitudinal EHR has received much attention in recent years. In this paper, we provide a comprehensive review of low-rank approximation-based approaches for computational phenotyping. The existing literature is categorized into temporal vs. static phenotyping approaches based on matrix vs. tensor decompositions. Furthermore, we outline different approaches for the validation of phenotypes, i.e., the assessment of clinical significance.

In this paper, we introduce a new task, spoken video grounding (SVG), which aims to localize the desired video fragments from spoken language descriptions. Compared with using text, employing audio requires the model to directly exploit the useful phonemes and syllables related to the video from raw speech. Moreover, we randomly add environmental noises to this speech audio, further increasing the difficulty of this task and better simulating real applications. To rectify the discriminative phonemes and extract video-related information from noisy audio, we develop a novel video-guided curriculum learning (VGCL) during the audio pre-training process, which can make use of the vital visual perceptions to help understand the spoken language and suppress the external noise. Considering during inference the model can not obtain ground truth video segments, we design a curriculum strategy that gradually shifts the input video from the ground truth to the entire video content during pre-training. Finally, the model can learn how to extract critical visual information from the entire video clip to help understand the spoken language. In addition, we collect the first large-scale spoken video grounding dataset based on ActivityNet, which is named as ActivityNet Speech dataset. Extensive experiments demonstrate our proposed video-guided curriculum learning can facilitate the pre-training process to obtain a mutual audio encoder, significantly promoting the performance of spoken video grounding tasks. Moreover, we prove that in the case of noisy sound, our model outperforms the method that grounding video with ASR transcripts, further demonstrating the effectiveness of our curriculum strategy.