This report describes our approach to design and evaluate a software stack for a race car capable of achieving competitive driving performance in the different disciplines of the Formula Student Driverless. By using a 360{\deg} LiDAR and optionally three cameras, we reliably recognize the plastic cones that mark the track boundaries at distances of around 35 m, enabling us to drive at the physical limits of the car. Using a GraphSLAM algorithm, we are able to map these cones with a root-mean-square error of less than 15 cm while driving at speeds of over 70 kph on a narrow track. The high-precision map is used in the trajectory planning to detect the lane boundaries using Delaunay triangulation and a parametric cubic spline. We calculate an optimized trajectory using a minimum curvature approach together with a GGS-diagram that takes the aerodynamics at different velocities into account. To track the target path with accelerations of up to 1.6 g, the control system is split into a PI controller for longitudinal control and model predictive controller for lateral control. Additionally, a low-level optimal control allocation is used. The software is realized in ROS C++ and tested in a custom simulation, as well as on the actual race track.

Detecting out-of-domain (OOD) intents from user queries is essential for avoiding wrong operations in task-oriented dialogue systems. The key challenge is how to distinguish in-domain (IND) and OOD intents. Previous methods ignore the alignment between representation learning and scoring function, limiting the OOD detection performance. In this paper, we propose a unified neighborhood learning framework (UniNL) to detect OOD intents. Specifically, we design a K-nearest neighbor contrastive learning (KNCL) objective for representation learning and introduce a KNN-based scoring function for OOD detection. We aim to align representation learning with scoring function. Experiments and analysis on two benchmark datasets show the effectiveness of our method.

Vision Transformers achieved outstanding performance in many computer vision tasks. Early Vision Transformers such as ViT and DeiT adopt global self-attention, which is computationally expensive when the number of patches is large. To improve efficiency, recent Vision Transformers adopt local self-attention mechanisms, where self-attention is computed within local windows. Despite the fact that window-based local self-attention significantly boosts efficiency, it fails to capture the relationships between distant but similar patches in the image plane. To overcome this limitation of image-space local attention, in this paper, we further exploit the locality of patches in the feature space. We group the patches into multiple clusters using their features, and self-attention is computed within every cluster. Such feature-space local attention effectively captures the connections between patches across different local windows but still relevant. We propose a Bilateral lOcal Attention vision Transformer (BOAT), which integrates feature-space local attention with image-space local attention. We further integrate BOAT with both Swin and CSWin models, and extensive experiments on several benchmark datasets demonstrate that our BOAT-CSWin model clearly and consistently outperforms existing state-of-the-art CNN models and vision Transformers.

Both in terrestrial and extraterrestrial environments, the precise and informative model of the ground and the surface ahead is crucial for navigation and obstacle avoidance. The ground surface is not always flat and it may be sloped, bumpy and rough specially in off-road terrestrial scenes. In bumpy and rough scenes the functional relationship of the surface-related features may vary in different areas of the ground, as the structure of the ground surface may vary suddenly and further the measured point cloud of the ground does not bear smoothness. Thus, the ground-related features must be obtained based on local estimates or even point estimates. To tackle this problem, the segment-wise GP-based ground segmentation method with local smoothness estimation is proposed. This method is an extension to our previous method in which a realistic measurement of the length-scale values were provided for the covariance kernel in each line-segment to give precise estimation of the ground for sloped terrains. In this extension, the value of the length-scale is estimated locally for each data point which makes it much more precise for the rough scenes while being not computationally complex and more robust to under-segmentation, sparsity and under-represent-ability. The segment-wise task is performed to estimate a partial continuous model of the ground for each radial range segment. Simulation results show the effectiveness of the proposed method to give a continuous and precise estimation of the ground surface in rough and bumpy scenes while being fast enough for real-world applications.

Volatility clustering is a common phenomenon in financial time series. Typically, linear models can be used to describe the temporal autocorrelation of the (logarithmic) variance of returns. Considering the difficulty in estimating this model, we construct a Dynamic Bayesian Network, which utilizes the conjugate prior relation of normal-gamma and gamma-gamma, so that its posterior form locally remains unchanged at each node. This makes it possible to find approximate solutions using variational methods quickly. Furthermore, we ensure that the volatility expressed by the model is an independent incremental process after inserting dummy gamma nodes between adjacent time steps. We have found that this model has two advantages: 1) It can be proved that it can express heavier tails than Gaussians, i.e., have positive excess kurtosis, compared to popular linear models. 2) If the variational inference(VI) is used for state estimation, it runs much faster than Monte Carlo(MC) methods since the calculation of the posterior uses only basic arithmetic operations. And its convergence process is deterministic. We tested the model, named Gam-Chain, using recent Crypto, Nasdaq, and Forex records of varying resolutions. The results show that: 1) In the same case of using MC, this model can achieve comparable state estimation results with the regular lognormal chain. 2) In the case of only using VI, this model can obtain accuracy that are slightly worse than MC, but still acceptable in practice; 3) Only using VI, the running time of Gam-Chain, in general case, can be reduced to below 5% of that based on the lognormal chain via MC.

Data-driven approaches and paradigms have become promising solutions to efficient network performances through optimization. These approaches focus on state-of-the-art machine learning techniques that can address the needs of 5G networks and the networks of tomorrow, such as proactive load balancing. In contrast to model-based approaches, data-driven approaches do not need accurate models to tackle the target problem, and their associated architectures provide a flexibility of available system parameters that improve the feasibility of learning-based algorithms in mobile wireless networks. The work presented in this paper focuses on demonstrating a working system prototype of the 5G Core (5GC) network and the Network Data Analytics Function (NWDAF) used to bring the benefits of data-driven techniques to fruition. Analyses of the network-generated data explore core intra-network interactions through unsupervised learning, clustering, and evaluate these results as insights for future opportunities and works.

Node embedding is a central topic in graph representation learning. Computational efficiency and scalability can be challenging to any method that requires full-graph operations. We propose sampling approaches to node embedding, with or without explicit modelling of the feature vector, which aim to extract useful information from both the eigenvectors related to the graph Laplacien and the given values associated with the graph.

Approximate Nearest Neighbor Search (ANNS) in high dimensional spaces is crucial for many real-life applications (e.g., e-commerce, web, multimedia, etc.) dealing with an abundance of data. This paper proposes an end-to-end learning framework that couples the partitioning (one critical step of ANNS) and learning-to-search steps using a custom loss function. A key advantage of our proposed solution is that it does not require any expensive pre-processing of the dataset, which is one of the critical limitations of the state-of-the-art approach. We achieve the above edge by formulating a multi-objective custom loss function that does not need ground truth labels to quantify the quality of a given data-space partition, making it entirely unsupervised. We also propose an ensembling technique by adding varying input weights to the loss function to train an ensemble of models to enhance the search quality. On several standard benchmarks for ANNS, we show that our method beats the state-of-the-art space partitioning method and the ubiquitous K-means clustering method while using fewer parameters and shorter offline training times. We also show that incorporating our space-partitioning strategy into state-of-the-art ANNS techniques such as ScaNN can improve their performance significantly. Finally, we present our unsupervised partitioning approach as a promising alternative to many widely used clustering methods, such as K-means clustering and DBSCAN.

Recent work shows that sensitive user data can be reconstructed from gradient updates, breaking the key privacy promise of federated learning. While success was demonstrated primarily on image data, these methods do not directly transfer to other domains such as text. In this work, we propose LAMP, a novel attack tailored to textual data, that successfully reconstructs original text from gradients. Our attack is based on two key insights: (i) modeling prior text probability with an auxiliary language model, guiding the search towards more natural text, and (ii) alternating continuous and discrete optimization, which minimizes reconstruction loss on embeddings, while avoiding local minima by applying discrete text transformations. Our experiments demonstrate that LAMP is significantly more effective than prior work: it reconstructs 5x more bigrams and 23% longer subsequences on average. Moreover, we are the first to recover inputs from batch sizes larger than 1 for textual models. These findings indicate that gradient updates of models operating on textual data leak more information than previously thought.

Abstract: Developing reliable methods to discriminate different transient brain states that change over time is a key neuroscientific challenge in brain imaging studies. Topological data analysis (TDA), a novel framework based on algebraic topology, can handle such a challenge. However, existing TDA has been somewhat limited to capturing the static summary of dynamically changing brain networks. We propose a novel dynamic-TDA framework that builds persistent homology over a time series of brain networks. We construct a Wasserstein distance based inference procedure to discriminate between time series of networks.