Untergeordnete Publikationstypen

Neueste Zugänge 

  1. AN ONTOLOGY-BASED REASONING FRAMEWORK 

    Pei, Wanyu; Xiong, Shuyan; Habert, Guillaume; et al. (2024)
    PROCEEDINGS OF THE 29TH INTERNATIONAL CONFERENCE OF THE ASSOCIATION FOR COMPUTER-AIDED ARCHITECTURAL DESIGN RESEARCH IN ASIA, CAADRIA 2024, VOL 2
    The materials stored in existing urban buildings represent a significant share of globally accumulated resources, the composition and quantity of which should be tracked for management and reuse purposes. Due to the coarse-grained nature of building data at the city level, the description of building material stock (BMS) is usually limited to the material intensity (MI) level of several key materials, omitting the component-level analysis ...
    Conference Paper
  2. UNCOVERING THE CIRCULAR POTENTIAL: ESTIMATING MATERIAL FLOWS FOR BUILDING SYSTEMS COMPONENTS REUSE IN THE SWISS BUILT ENVIRONMENT 

    Xiong, Shuyan; Escamilla, Edwin Zea; Habert, Guillaume (2024)
    PROCEEDINGS OF THE 29TH INTERNATIONAL CONFERENCE OF THE ASSOCIATION FOR COMPUTER-AIDED ARCHITECTURAL DESIGN RESEARCH IN ASIA, CAADRIA 2024, VOL 1
    The construction industry plays a critical role in global resource consumption and greenhouse gas emissions, highlighting the urgent need for sustainable development practices. However, a key challenge in this area is the lack of effective models for resource use that align with circular economy principles. This gap hinders efforts to achieve sustainable resource management, especially in the face of increasing urbanization and material ...
    Conference Paper
  3. Sequential Deconfounding for Causal Inference with Unobserved Confounders 

    Hatt, Tobias; Feuerriegel, Stefan (2024)
    Proceedings of Machine Learning Research
    Observational data is often used to estimate the effect of a treatment when randomized experiments are infeasible or costly. However, observational data often yields biased estimates of treatment effects, since treatment assignment can be confounded by unobserved variables. A remedy is offered by deconfounding methods that adjust for such unobserved confounders. In this paper, we develop the Sequential Deconfounder, a method that enables ...
    Conference Paper
  4. On the use of seashells as green solution to mechanically stabilise dredged sediments 

    Petti R.; Vitone C.; Marchi M.I.; et al. (2024)
    E3S Web of Conferences
    The article reports the results of an experimental activity conducted on dredged fine-grained marine sediments and aimed to find out novel eco-friendly solutions for their mechanical stabilisation. The main idea of this research is to use seashells, i.e., another waste material, to partially replace cement binders in the mechanical stabilisation of sediments for the production of a new stable material that can potentially be used in ...
    Conference Paper
  5. Experimental evidences of bio-chemo-mechanical processes in contaminated sediments 

    Sollecito F.; Plötze M.; Puzrin A.M.; et al. (2024)
    E3S Web of Conferences
    The research shows the results of a micro to macro testing programme carried out on contaminated marine sediments from a natural deposit to assess the effects of bio-chemo-mechanical coupled processes which may act in complex natural environments and affect the geotechnical properties of the clays. The research has been triggered by the emblematic case of the contaminated Mar Piccolo (MP) basin in Taranto (Southern Italy), where the high ...
    Conference Paper
  6. CO2-neutraler Prozessdampf für die Industrie: Analyse von Technologien und Strategien 

    Roskosch, Dennis; Nolzen, Niklas; Evering, Luisa; et al. (2024)
    Other Conference Item
  7. Tight bounds for maximum <i>l</i><sub>1</sub>-margin classifiers 

    Stojanovic, Stefan; Donhauser, Konstantin; Yang, Fanny (2024)
    Proceedings of Machine Learning Research
    Popular iterative algorithms such as boosting methods and coordinate descent on linear models converge to the maximum l(1)-margin classifier, a.k.a. sparse hard-margin SVM, in high dimensional regimes where the data is linearly separable. Previous works consistently show that many estimators relying on the l(1)-norm achieve improved statistical rates for hard sparse ground truths. We show that surprisingly, this adaptivity does not apply ...
    Conference Paper
  8. Boosting Verification of Deep Reinforcement Learning via Piece-wise Linear Decision Neural Networks 

    Tian, Jiaxu; Zhi, Dapeng; Liu, Si; et al. (2023)
    Advances in Neural Information Processing Systems
    Formally verifying deep reinforcement learning (DRL) systems suffers from both inaccurate verification results and limited scalability. The major obstacle lies in the large overestimation introduced inherently during training and then transforming the inexplicable decision-making models i.e., deep neural networks (DNNs), into easy-to-verify models. In this paper, we propose an inverse transform-then-train approach, which first encodes a ...
    Conference Paper
  9. Can semi-supervised learning use all the data effectively? A lower bound perspective 

    Tifrea, Alexandru; Yuce, Gizem; Sanyal, Amartya; et al. (2023)
    Advances in Neural Information Processing Systems
    Prior theoretical and empirical works have established that semi-supervised learning algorithms can leverage the unlabeled data to improve over the labeled sample complexity of supervised learning (SL) algorithms. However, existing theoretical work focuses on regimes where the unlabeled data is sufficient to learn a good decision boundary using unsupervised learning (UL) alone. This begs the question: Can SSL algorithms simultaneously ...
    Conference Paper
  10. DataPerf: Benchmarks for Data-Centric AI Development 

    Mazumder, Mark; Banbury, Colby; Yao, Xiaozhe; et al. (2023)
    Advances in Neural Information Processing Systems
    Machine learning research has long focused on models rather than datasets, and prominent datasets are used for common ML tasks without regard to the breadth, difficulty, and faithfulness of the underlying problems. Neglecting the fundamental importance of data has given rise to inaccuracy, bias, and fragility in real-world applications, and research is hindered by saturation across existing dataset benchmarks. In response, we present ...
    Conference Paper
  11. The apparent decoupling of magmatic and hydrothermal activities in the Chuquicamata District 

    Virmond, Adrianna L.; Selby, David; Szymanowski, Dawid; et al. (2023)
    17TH BIENNIAL SGA MEETING, 2023, VOL 1
    Understanding the primary controls on the formation of mineral deposits is fundamental to assist exploration geologists in finding more resources in an ever growing high-demand world. Even though Porphyry Copper Systems have been extensively researched, the primary controls on the tonnage of deposits are still poorly quantified. Recent studies suggest the timescales and physical parameters involved in the formation of such deposits might ...
    Conference Paper
  12. Estimating Generic 3D Room Structures from 2D Annotations 

    Rozumnyi, Denys; Popov, Stefan; Maninis, Kevis-Kokitsi; et al. (2023)
    Advances in Neural Information Processing Systems
    Indoor rooms are among the most common use cases in 3D scene understanding. Current state-of-the-art methods for this task are driven by large annotated datasets. Room layouts are especially important, consisting of structural elements in 3D, such as wall, floor, and ceiling. However, they are difficult to annotate, especially on pure RGB video. We propose a novel method to produce generic 3D room layouts just from 2D segmentation masks, ...
    Conference Paper
  13. CLADDER: Assessing Causal Reasoning in Language Models 

    Jin, Zhijing; Chen, Yuen; Leeb, Felix; et al. (2023)
    Advances in Neural Information Processing Systems 36
    The ability to perform causal reasoning is widely considered a core feature of intelligence. In this work, we investigate whether large language models (LLMs) can coherently reason about causality. Much of the existing work in natural language processing (NLP) focuses on evaluating commonsense causal reasoning in LLMs, thus failing to assess whether a model can perform causal inference in accordance with a set of well-defined formal rules. ...
    Conference Paper
  14. Convolutional Neural Operators for robust and accurate learning of PDEs 

    Raonic, Bogdan; Molinaro, Roberto; De Ryck, Tim; et al. (2023)
    Advances in Neural Information Processing Systems 36
    Although very successfully used in conventional machine learning, convolution based neural network architectures - believed to be inconsistent in function space - have been largely ignored in the context of learning solution operators of PDEs. Here, we present novel adaptations for convolutional neural networks to demonstrate that they are indeed able to process functions as inputs and outputs. The resulting architecture, termed as ...
    Conference Paper
  15. Counterfactual Memorization in Neural Language Models 

    Zhang, Chiyuan; Ippolito, Daphne; Lee, Katherine; et al. (2023)
    Advances in Neural Information Processing Systems 36
    Modern neural language models that are widely used in various NLP tasks risk memorizing sensitive information from their training data. Understanding this memorization is important in real world applications and also from a learning-theoretical perspective. An open question in previous studies of language model memorization is how to filter out "common" memorization. In fact, most memorization criteria strongly correlate with the number ...
    Conference Paper
  16. Explore In-Context Learning for 3D Point Cloud Understanding 

    Fang, Zhongbin; Li, Xiangtai; Li, Xia; et al. (2023)
    Advances in Neural Information Processing Systems 36
    With the rise of large-scale models trained on broad data, in-context learning has become a new learning paradigm that has demonstrated significant potential in natural language processing and computer vision tasks. Meanwhile, in-context learning is still largely unexplored in the 3D point cloud domain. Although masked modeling has been successfully applied for in-context learning in 2D vision, directly extending it to 3D point clouds ...
    Conference Paper
  17. DISCO-10M: A Large-Scale Music Dataset 

    Lanzendörfer, Luca; Grötschla, Florian; Funke, Emil; et al. (2023)
    Advances in Neural Information Processing Systems 36
    Music datasets play a crucial role in advancing research in machine learning for music. However, existing music datasets suffer from limited size, accessibility, and lack of audio resources. To address these shortcomings, we present DISCO-10M, a novel and extensive music dataset that surpasses the largest previously available music dataset by an order of magnitude. To ensure high-quality data, we implement a multi-stage filtering process. ...
    Conference Paper
  18. Contextual Stochastic Bilevel Optimization 

    Hu, Yifan; Wang, Jie; Xie, Yao; et al. (2023)
    Advances in Neural Information Processing Systems 36
    We introduce contextual stochastic bilevel optimization (CSBO) - a stochastic bilevel optimization framework with the lower-level problem minimizing an expectation conditioned on some contextual information and the upper-level decision variable. This framework extends classical stochastic bilevel optimization when the lower-level decision maker responds optimally not only to the decision of the upper-level decision maker but also to some ...
    Conference Paper
  19. Canonical normalizing flows for manifold learning 

    Flouris, Kyriakos; Konukoglu, Ender (2023)
    Advances in Neural Information Processing Systems 36
    Manifold learning flows are a class of generative modelling techniques that assume a low-dimensional manifold description of the data. The embedding of such a manifold into the high-dimensional space of the data is achieved via learnable invertible transformations. Therefore, once the manifold is properly aligned via a reconstruction loss, the probability density is tractable on the manifold and maximum likelihood can be used to optimize ...
    Conference Paper
  20. Estimating the Rate-Distortion Function by Wasserstein Gradient Descent 

    Yang, Yibo; Eckstein, Stephan; Nutz, Marcel; et al. (2023)
    Advances in Neural Information Processing Systems 36
    In the theory of lossy compression, the rate-distortion (R-D) function R(D) describes how much a data source can be compressed (in bit-rate) at any given level of fidelity (distortion). Obtaining R(D) for a given data source establishes the fundamental performance limit for all compression algorithms. We propose a new method to estimate R(D) from the perspective of optimal transport. Unlike the classic Blahut-Arimoto algorithm which fixes ...
    Conference Paper

Mehr anzeigen