Abstract
Despite recent progress in improving the performance of misinformation detection systems, classifying misinformation in an unseen domain remains an elusive challenge. To address this issue, a common approach is to introduce a domain critic and encourage domain-invariant input features. However, early misinformation often demonstrates both conditional and label shifts against existing misinformation data (e.g., class imbalance in COVID-19 datasets), rendering such methods less effective for detecting early misinformation. In this paper, we propose contrastive adaptation network for early misinformation detection (CANMD). Specifically, we leverage pseudo labeling to generate high-confidence target examples for joint training with source data. We additionally design a label correction component to estimate and correct the label shifts (i.e., class priors) between the source and target domains. Moreover, a contrastive adaptation loss is integrated in the objective function to reduce the intra-class discrepancy and enlarge the inter-class discrepancy. As such, the adapted model learns corrected class priors and an invariant conditional distribution across both domains for improved estimation of the target data distribution. To demonstrate the effectiveness of the proposed CANMD, we study the case of COVID-19 early misinformation detection and perform extensive experiments using multiple real-world datasets. The results suggest that CANMD can effectively adapt misinformation detection systems to the unseen COVID-19 target domain with significant improvements compared to the state-of-the-art baselines.
Abstract
Image clustering is an important, and open challenge task in computer vision. Although many methods have been proposed to solve the image clustering task, they only explore images and uncover clusters according to the image features, thus are unable to distinguish visually similar but semantically different images. In this paper, we propose to investigate the task of image clustering with the help of visual-language pre-training model. Different from the zero-shot setting in which the class names are known, we only know the number of clusters in this setting. Therefore, how to map images to a proper semantic space and how to cluster images from both image and semantic spaces are two key problems. To solve the above problems, we propose a novel image clustering method guided by the visual-language pre-training model CLIP, named as \textbf{Semantic-enhanced Image Clustering (SIC)}. In this new method, we propose a method to map the given images to a proper semantic space first and efficient methods to generate pseudo-labels according to the relationships between images and semantics. Finally, we propose to perform clustering with the consistency learning in both image space and semantic space, in a self-supervised learning fashion. Theoretical result on convergence analysis shows that our proposed method can converge in sublinear speed. Theoretical analysis on expectation risk also shows that we can reduce the expectation risk by improving the neighborhood consistency or prediction confidence or reducing neighborhood imbalance. Experimental results on five benchmark datasets clearly show the superiority of our new method.
PLMCL: Partial-Label Momentum Curriculum Learning for Multi-Label Image Classification
Authors: Rabab Abdelfattah, Xin Zhang, Zhenyao Wu, Xinyi Wu, Xiaofeng Wang, Song Wang
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Multi-label image classification aims to predict all possible labels in an image. It is usually formulated as a partial-label learning problem, given the fact that it could be expensive in practice to annotate all labels in every training image. Existing works on partial-label learning focus on the case where each training image is annotated with only a subset of its labels. A special case is to annotate only one positive label in each training image. To further relieve the annotation burden and enhance the performance of the classifier, this paper proposes a new partial-label setting in which only a subset of the training images are labeled, each with only one positive label, while the rest of the training images remain unlabeled. To handle this new setting, we propose an end-to-end deep network, PLMCL (Partial Label Momentum Curriculum Learning), that can learn to produce confident pseudo labels for both partially-labeled and unlabeled training images. The novel momentum-based law updates soft pseudo labels on each training image with the consideration of the updating velocity of pseudo labels, which help avoid trapping to low-confidence local minimum, especially at the early stage of training in lack of both observed labels and confidence on pseudo labels. In addition, we present a confidence-aware scheduler to adaptively perform easy-to-hard learning for different labels. Extensive experiments demonstrate that our proposed PLMCL outperforms many state-of-the-art multi-label classification methods under various partial-label settings on three different datasets.
Minkowski Tracker: A Sparse Spatio-Temporal R-CNN for Joint Object Detection and Tracking
Abstract
Recent research in multi-task learning reveals the benefit of solving related problems in a single neural network. 3D object detection and multi-object tracking (MOT) are two heavily intertwined problems predicting and associating an object instance location across time. However, most previous works in 3D MOT treat the detector as a preceding separated pipeline, disjointly taking the output of the detector as an input to the tracker. In this work, we present Minkowski Tracker, a sparse spatio-temporal R-CNN that jointly solves object detection and tracking. Inspired by region-based CNN (R-CNN), we propose to solve tracking as a second stage of the object detector R-CNN that predicts assignment probability to tracks. First, Minkowski Tracker takes 4D point clouds as input to generate a spatio-temporal Bird's-eye-view (BEV) feature map through a 4D sparse convolutional encoder network. Then, our proposed TrackAlign aggregates the track region-of-interest (ROI) features from the BEV features. Finally, Minkowski Tracker updates the track and its confidence score based on the detection-to-track match probability predicted from the ROI features. We show in large-scale experiments that the overall performance gain of our method is due to four factors: 1. The temporal reasoning of the 4D encoder improves the detection performance 2. The multi-task learning of object detection and MOT jointly enhances each other 3. The detection-to-track match score learns implicit motion model to enhance track assignment 4. The detection-to-track match score improves the quality of the track confidence score. As a result, Minkowski Tracker achieved the state-of-the-art performance on Nuscenes dataset tracking task without hand-designed motion models.
A Trust Framework for Government Use of Artificial Intelligence and Automated Decision Making
Authors: Pia Andrews, Tim de Sousa, Bruce Haefele, Matt Beard, Marcus Wigan, Abhinav Palia, Kathy Reid, Saket Narayan, Morgan Dumitru, Alex Morrison, Geoff Mason, Aurelie Jacquet
Abstract
This paper identifies the current challenges of the mechanisation, digitisation and automation of public sector systems and processes, and proposes a modern and practical framework to ensure and assure ethical and high veracity Artificial Intelligence (AI) or Automated Decision Making (ADM) systems in public institutions. This framework is designed for the specific context of the public sector, in the jurisdictional and constitutional context of Australia, but is extendable to other jurisdictions and private sectors. The goals of the framework are to: 1) earn public trust and grow public confidence in government systems; 2) to ensure the unique responsibilities and accountabilities (including to the public) of public institutions under Administrative Law are met effectively; and 3) to assure a positive human, societal and ethical impact from the adoption of such systems. The framework could be extended to assure positive environmental or other impacts, but this paper focuses on human/societal outcomes and public trust. This paper is meant to complement principles-based frameworks like Australia's Artificial Intelligence Ethics Framework and the EU Assessment List for Trustworthy AI. In many countries, COVID created a bubble of improved trust, a bubble which has arguably already popped, and in an era of unprecedented mistrust of public institutions (but even in times of high trust) it is not enough that a service is faster, or more cost-effective. This paper proposes recommendations for government systems (technology platforms, operations, culture, governance, engagement, etc.) that would help to improve public confidence and trust in public institutions, policies and services, whilst meeting the special obligations and responsibilities of the public sector.
Towards Confidence-aware Calibrated Recommendation
Authors: Mohammadmehdi Naghiaei, Hossein A. Rahmani, Mohammad Aliannejadi, Nasim Sonboli
Abstract
Recommender systems utilize users' historical data to learn and predict their future interests, providing them with suggestions tailored to their tastes. Calibration ensures that the distribution of recommended item categories is consistent with the user's historical data. Mitigating miscalibration brings various benefits to a recommender system. For example, it becomes less likely that a system overlooks categories with less interaction on a user's profile by only recommending popular categories. Despite the notable success, calibration methods have several drawbacks, such as limiting the diversity of the recommended items and not considering the calibration confidence. This work, presents a set of properties that address various aspects of a desired calibrated recommender system. Considering these properties, we propose a confidence-aware optimization-based re-ranking algorithm to find the balance between calibration, relevance, and item diversity, while simultaneously accounting for calibration confidence based on user profile size. Our model outperforms state-of-the-art methods in terms of various accuracy and beyond-accuracy metrics for different user groups.
Semi-supervised classification using a supervised autoencoder for biomedical applications
Authors: Cyprien Gille, Frederic Guyard, Michel Barlaud
Abstract
In this paper we present a new approach to solve semi-supervised classification tasks for biomedical applications, involving a supervised autoencoder network. We create a network architecture that encodes labels into the latent space of an autoencoder, and define a global criterion combining classification and reconstruction losses. We train the Semi-Supervised AutoEncoder (SSAE) on labelled data using a double descent algorithm. Then, we classify unlabelled samples using the learned network thanks to a softmax classifier applied to the latent space which provides a classification confidence score for each class. We implemented our SSAE method using the PyTorch framework for the model, optimizer, schedulers, and loss functions. We compare our semi-supervised autoencoder method (SSAE) with classical semi-supervised methods such as Label Propagation and Label Spreading, and with a Fully Connected Neural Network (FCNN). Experiments show that the SSAE outperforms Label Propagation and Spreading and the Fully Connected Neural Network both on a synthetic dataset and on two real-world biological datasets.
Simulation-Informed Revenue Extrapolation with Confidence Estimate for Scaleup Companies Using Scarce Time-Series Data
Authors: Lele Cao, Sonja Horn, Vilhelm von Ehrenheim, Richard Anselmo Stahl, Henrik Landgren
Subjects: Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
Abstract
Investment professionals rely on extrapolating company revenue into the future (i.e. revenue forecast) to approximate the valuation of scaleups (private companies in a high-growth stage) and inform their investment decision. This task is manual and empirical, leaving the forecast quality heavily dependent on the investment professionals' experiences and insights. Furthermore, financial data on scaleups is typically proprietary, costly and scarce, ruling out the wide adoption of data-driven approaches. To this end, we propose a simulation-informed revenue extrapolation (SiRE) algorithm that generates fine-grained long-term revenue predictions on small datasets and short time-series. SiRE models the revenue dynamics as a linear dynamical system (LDS), which is solved using the EM algorithm. The main innovation lies in how the noisy revenue measurements are obtained during training and inferencing. SiRE works for scaleups that operate in various sectors and provides confidence estimates. The quantitative experiments on two practical tasks show that SiRE significantly surpasses the baseline methods by a large margin. We also observe high performance when SiRE extrapolates from short time-series and predicts for long-term. The performance-efficiency balance and result explainability of SiRE are also validated empirically. Evaluated from the perspective of investment professionals, SiRE can precisely locate the scaleups that have a great potential return in 2 to 5 years. Furthermore, our qualitative inspection illustrates some advantageous attributes of the SiRE revenue forecasts.
Verification-Preserving Inlining in Automatic Separation Logic Verifiers (extended version)
Authors: Thibault Dardinier, Gaurav Parthasarathy, Peter Müller
Abstract
Bounded verification has proved useful to detect bugs and to increase confidence in the correctness of a program. In contrast to unbounded verification, reasoning about calls via (bounded) inlining and about loops via (bounded) unrolling does not require method specifications and loop invariants and, therefore, reduces the annotation overhead to the bare minimum, namely specifications of the properties to be verified. For verifiers based on traditional program logics, verification via inlining (and unrolling) is verification-preserving: successful unbounded verification of a program w.r.t. some annotation implies successful verification of the inlined program. That is, any error detected in the inlined program reveals a true error in the original program. However, this essential property might not hold for automatic separation logic verifiers such as Caper, GRASShopper, RefinedC, Steel, VeriFast, and verifiers based on Viper. In this setting, inlining generally changes the resources owned by method executions, which may affect automatic proof search algorithms and introduce spurious errors. In this paper, we present the first technique for verification-preserving inlining in automatic separation logic verifiers. We identify a semantic condition on programs and prove in Isabelle/HOL that it ensures verification-preserving inlining for state-of-the-art automatic separation logic verifiers. We also prove a dual result: successful verification of the inlined program ensures that there are method and loop annotations that enable the verification of the original program for bounded executions. To check our semantic condition automatically, we present two approximations that can be checked syntactically and with a program verifier, respectively. We implemented these checks in Viper and demonstrate that they are effective for non-trivial examples from different verifiers.
Keyword: scaling
Phase Shift-Free Passive Beamforming for Reconfigurable Intelligent Surfaces
Authors: Aymen Khaleel, Ertugrul Basar
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
Reconfigurable intelligent surface (RIS)-assisted communications recently appeared as a game-changing technology for next-generation wireless communications due to its unprecedented ability to reform the propagation environment. One of the main aspects of using RISs is the exploitation of the so-called passive beamforming (PB), which is carried out by adjusting the reflection coefficients (mainly the phase shifts) of the individual RIS elements. However, practically, this individual phase shift adjustment is associated with many issues in hardware implementation, limiting the RIS achievable gain. In this paper, we propose a low-cost, phase shift-free and novel PB scheme by only optimizing the on/off states of the RIS elements while fixing their phase shifts. The proposed PB scheme is shown to achieve the same scaling law (quadratic growth with the RIS size) for the signal-to-noise ratio as in the classical phase shift-based PB scheme, yet, with far less sensitivity to spatial correlation and phase errors. We provide a unified mathematical analysis that characterizes the performance of the proposed PB scheme and obtain the outage probability for the considered RIS-assisted system. Based on the provided computer simulations, the proposed PB scheme is shown to have a clear superiority over the classical one under different performance metrics.
zk-PCN: A Privacy-Preserving Payment Channel Network Using zk-SNARKs
Abstract
Payment channel network (PCN) is a layer-two scaling solution that enables fast off-chain transactions but does not involve on-chain transaction settlement. PCNs raise new privacy issues including balance secrecy, relationship anonymity and payment privacy. Moreover, protecting privacy causes low transaction success rates. To address this dilemma, we propose zk-PCN, a privacy-preserving payment channel network using zk-SNARKs. We prevent from exposing true balances by setting up \textit{public balances} instead. Using public balances, zk-PCN can guarantee high transaction success rates and protect PCN privacy with zero-knowledge proofs. Additionally, zk-PCN is compatible with the existing routing algorithms of PCNs. To support such compatibility, we propose zk-IPCN to improve zk-PCN with a novel proof generation (RPG) algorithm. zk-IPCN reduces the overheads of storing channel information and lowers the frequency of generating zero-knowledge proofs. Finally, extensive simulations demonstrate the effectiveness and efficiency of zk-PCN in various settings.
Scalable link prediction in Twitter using self-configured framework
Authors: Nur Nasuha Daud, Siti Hafizah Ab Hamid, Chempaka Seri, Muntadher Saadoon, Nor Badrul Anuar
Abstract
Link prediction analysis becomes vital to acquire a deeper understanding of events underlying social networks interactions and connections especially in current evolving and large-scale social networks. Traditional link prediction approaches underperformed for most large-scale social networks in terms of its scalability and efficiency. Spark is a distributed open-source framework that facilitate scalable link prediction efficiency in large-scale social networks. The framework provides numerous tunable properties for users to manually configure the parameters for the applications. However, manual configurations open to performance issue when the applications start scaling tremendously, which is hard to set up and expose to human errors. This paper introduced a novel Self-Configured Framework (SCF) to provide an autonomous feature in Spark that predicts and sets the best configuration instantly before the application execution using XGBoost classifier. SCF is evaluated on the Twitter social network using three link prediction applications: Graph Clustering (GC), Overlapping Community Detection (OCD), and Redundant Graph Clustering (RGD) to assess the impact of shifting data sizes on different applications in Twitter. The result demonstrates a 40% reduction in prediction time as well as a balanced resource consumption that makes full use of resources, especially for limited number and size of clusters
SIM2E: Benchmarking the Group Equivariant Capability of Correspondence Matching Algorithms
Authors: Shuai Su, Zhongkai Zhao, Yixin Fei, Shuda Li, Qijun Chen, Rui Fan
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Correspondence matching is a fundamental problem in computer vision and robotics applications. Solving correspondence matching problems using neural networks has been on the rise recently. Rotation-equivariance and scale-equivariance are both critical in correspondence matching applications. Classical correspondence matching approaches are designed to withstand scaling and rotation transformations. However, the features extracted using convolutional neural networks (CNNs) are only translation-equivariant to a certain extent. Recently, researchers have strived to improve the rotation-equivariance of CNNs based on group theories. Sim(2) is the group of similarity transformations in the 2D plane. This paper presents a specialized dataset dedicated to evaluating sim(2)-equivariant correspondence matching algorithms. We compare the performance of 16 state-of-the-art (SoTA) correspondence matching approaches. The experimental results demonstrate the importance of group equivariant algorithms for correspondence matching on various sim(2) transformation conditions. Since the subpixel accuracy achieved by CNN-based correspondence matching approaches is unsatisfactory, this specific area requires more attention in future works. Our dataset is publicly available at: mias.group/SIM2E.
On uniqueness and stable estimation of multiple parameters in the Cahn-Hilliard equation
Authors: Aaron Brunk, Herbert Egger, Oliver Habrich
Abstract
We consider the identifiability and stable numerical estimation of multiple parameters in a Cahn-Hilliard model for phase separation. Spatially resolved measurements of the phase fraction are assumed to be accessible, with which the identifiability of single and multiple parameters up to certain scaling invariances is established. A regularized equation error approach is proposed for the stable numerical solution of the parameter identification problems, and convergence of the regularized approximations is proven under reasonable assumptions on the data noise. The viability of the theoretical results and the proposed methods is demonstrated in numerical tests.
Efficient Planning in a Compact Latent Action Space
Authors: Zhengyao Jiang, Tianjun Zhang, Michael Janner, Yueying Li, Tim Rocktäschel, Edward Grefenstette, Yuandong Tian
Abstract
While planning-based sequence modelling methods have shown great potential in continuous control, scaling them to high-dimensional state-action sequences remains an open challenge due to the high computational complexity and innate difficulty of planning in high-dimensional spaces. We propose the Trajectory Autoencoding Planner (TAP), a planning-based sequence modelling RL method that scales to high state-action dimensionalities. Using a state-conditional Vector-Quantized Variational Autoencoder (VQ-VAE), TAP models the conditional distribution of the trajectories given the current state. When deployed as an RL agent, TAP avoids planning step-by-step in a high-dimensional continuous action space but instead looks for the optimal latent code sequences by beam search. Unlike $O(D^3)$ complexity of Trajectory Transformer, TAP enjoys constant $O(C)$ planning computational complexity regarding state-action dimensionality $D$. Our empirical evaluation also shows the increasingly strong performance of TAP with the growing dimensionality. For Adroit robotic hand manipulation tasks with high state and action dimensionality, TAP surpasses existing model-based methods, including TT, with a large margin and also beats strong model-free actor-critic baselines.
FORBID: Fast Overlap Removal By stochastic gradIent Descent for Graph Drawing
Abstract
While many graph drawing algorithms consider nodes as points, graph visualization tools often represent them as shapes. These shapes support the display of information such as labels or encode various data with size or color. However, they can create overlaps between nodes which hinder the exploration process by hiding parts of the information. It is therefore of utmost importance to remove these overlaps to improve graph visualization readability. If not handled by the layout process, Overlap Removal (OR) algorithms have been proposed as layout post-processing. As graph layouts usually convey information about their topology, it is important that OR algorithms preserve them as much as possible. We propose a novel algorithm that models OR as a joint stress and scaling optimization problem, and leverages efficient stochastic gradient descent. This approach is compared with state-of-the-art algorithms, and several quality metrics demonstrate its efficiency to quickly remove overlaps while retaining the initial layout structures.
Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks
Abstract
A big convergence of language, vision, and multimodal pretraining is emerging. In this work, we introduce a general-purpose multimodal foundation model BEiT-3, which achieves state-of-the-art transfer performance on both vision and vision-language tasks. Specifically, we advance the big convergence from three aspects: backbone architecture, pretraining task, and model scaling up. We introduce Multiway Transformers for general-purpose modeling, where the modular architecture enables both deep fusion and modality-specific encoding. Based on the shared backbone, we perform masked "language" modeling on images (Imglish), texts (English), and image-text pairs ("parallel sentences") in a unified manner. Experimental results show that BEiT-3 obtains state-of-the-art performance on object detection (COCO), semantic segmentation (ADE20K), image classification (ImageNet), visual reasoning (NLVR2), visual question answering (VQAv2), image captioning (COCO), and cross-modal retrieval (Flickr30K, COCO).
Keyword: calibration
Hilti-Oxford Dataset: A Millimetre-Accurate Benchmark for Simultaneous Localization and Mapping
Authors: Lintong Zhang, Michael Helmberger, Lanke Frank Tarimo Fu, David Wisth, Marco Camurri, Davide Scaramuzza, Maurice Fallon
Subjects: Robotics (cs.RO); Image and Video Processing (eess.IV)
Abstract
Simultaneous Localization and Mapping (SLAM) is being deployed in real-world applications, however many state-of-the-art solutions still struggle in many common scenarios. A key necessity in progressing SLAM research is the availability of high-quality datasets and fair and transparent benchmarking. To this end, we have created the Hilti-Oxford Dataset, to push state-of-the-art SLAM systems to their limits. The dataset has a variety of challenges ranging from sparse and regular construction sites to a 17th century neoclassical building with fine details and curved surfaces. To encourage multi-modal SLAM approaches, we designed a data collection platform featuring a lidar, five cameras, and an IMU (Inertial Measurement Unit). With the goal of benchmarking SLAM algorithms for tasks where accuracy and robustness are paramount, we implemented a novel ground truth collection method that enables our dataset to accurately measure SLAM pose errors with millimeter accuracy. To further ensure accuracy, the extrinsics of our platform were verified with a micrometer-accurate scanner, and temporal calibration was managed online using hardware time synchronization. The multi-modality and diversity of our dataset attracted a large field of academic and industrial researchers to enter the second edition of the Hilti SLAM challenge, which concluded in June 2022. The results of the challenge show that while the top three teams could achieve accuracy of 2cm or better for some sequences, the performance dropped off in more difficult sequences.
Towards Calibrated Hyper-Sphere Representation via Distribution Overlap Coefficient for Long-tailed Learning
Abstract
Long-tailed learning aims to tackle the crucial challenge that head classes dominate the training procedure under severe class imbalance in real-world scenarios. However, little attention has been given to how to quantify the dominance severity of head classes in the representation space. Motivated by this, we generalize the cosine-based classifiers to a von Mises-Fisher (vMF) mixture model, denoted as vMF classifier, which enables to quantitatively measure representation quality upon the hyper-sphere space via calculating distribution overlap coefficient. To our knowledge, this is the first work to measure representation quality of classifiers and features from the perspective of distribution overlap coefficient. On top of it, we formulate the inter-class discrepancy and class-feature consistency loss terms to alleviate the interference among the classifier weights and align features with classifier weights. Furthermore, a novel post-training calibration algorithm is devised to zero-costly boost the performance via inter-class overlap coefficients. Our method outperforms previous work with a large margin and achieves state-of-the-art performance on long-tailed image classification, semantic segmentation, and instance segmentation tasks (e.g., we achieve 55.0\% overall accuracy with ResNetXt-50 in ImageNet-LT). Our code is available at https://github.com/VipaiLab/vMF\_OP.
Towards Confidence-aware Calibrated Recommendation
Authors: Mohammadmehdi Naghiaei, Hossein A. Rahmani, Mohammad Aliannejadi, Nasim Sonboli
Abstract
Recommender systems utilize users' historical data to learn and predict their future interests, providing them with suggestions tailored to their tastes. Calibration ensures that the distribution of recommended item categories is consistent with the user's historical data. Mitigating miscalibration brings various benefits to a recommender system. For example, it becomes less likely that a system overlooks categories with less interaction on a user's profile by only recommending popular categories. Despite the notable success, calibration methods have several drawbacks, such as limiting the diversity of the recommended items and not considering the calibration confidence. This work, presents a set of properties that address various aspects of a desired calibrated recommender system. Considering these properties, we propose a confidence-aware optimization-based re-ranking algorithm to find the balance between calibration, relevance, and item diversity, while simultaneously accounting for calibration confidence based on user profile size. Our model outperforms state-of-the-art methods in terms of various accuracy and beyond-accuracy metrics for different user groups.
Keyword: out of distribution detection
There is no result
Keyword: out-of-distribution detection
There is no result
Keyword: expected calibration error
There is no result
Keyword: overconfident
There is no result
Keyword: overconfidence
There is no result
Keyword: confidence
Contrastive Domain Adaptation for Early Misinformation Detection: A Case Study on COVID-19
Semantic-enhanced Image Clustering
PLMCL: Partial-Label Momentum Curriculum Learning for Multi-Label Image Classification
Minkowski Tracker: A Sparse Spatio-Temporal R-CNN for Joint Object Detection and Tracking
A Trust Framework for Government Use of Artificial Intelligence and Automated Decision Making
Towards Confidence-aware Calibrated Recommendation
Semi-supervised classification using a supervised autoencoder for biomedical applications
Simulation-Informed Revenue Extrapolation with Confidence Estimate for Scaleup Companies Using Scarce Time-Series Data
Verification-Preserving Inlining in Automatic Separation Logic Verifiers (extended version)
Keyword: scaling
Phase Shift-Free Passive Beamforming for Reconfigurable Intelligent Surfaces
zk-PCN: A Privacy-Preserving Payment Channel Network Using zk-SNARKs
Scalable link prediction in Twitter using self-configured framework
SIM2E: Benchmarking the Group Equivariant Capability of Correspondence Matching Algorithms
On uniqueness and stable estimation of multiple parameters in the Cahn-Hilliard equation
Efficient Planning in a Compact Latent Action Space
FORBID: Fast Overlap Removal By stochastic gradIent Descent for Graph Drawing
Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks
Keyword: calibration
Hilti-Oxford Dataset: A Millimetre-Accurate Benchmark for Simultaneous Localization and Mapping
Towards Calibrated Hyper-Sphere Representation via Distribution Overlap Coefficient for Long-tailed Learning
Towards Confidence-aware Calibrated Recommendation