arXiv subjects

Michael Schmidt

Publications and source records attributed to Michael Schmidt.

At least 37 records · Page 2Linked to original sources

arXiv2024

Remote Analysis of Femoroacetabular Impingement by a triade of label-free optical spectroscopy techniques

This paper introduces the combination of Laser-induced breakdown spectroscopy (LIBS), Raman spectroscopy (RS) and diffuse reflectance spectroscopy (DRS) in the field of biomedical research. Thereby, the results from RS and LIBS are combined with previous DRS results. These noninvasive optical methods, used together, offer thorough analysis of the absorptive and scattering behaviour, elemental composition, and molecular bindings of tissue, without resulting in considerable harm or changes to the sample or the requirement of markers. The study centres on applying this triad of techniques to tissues affected by Femoroacetabular Impingement (FAI), a condition characterised by reduced hip mobility due to developmental deformities of the hip joint. The research results in a biochemical pathway model of the condition causing the red staining caused by FAI which origin was unknown until now. This proves that this approach may have significant implications for numerous medical applications and support in exploring complex chemical and biochemical pathways.

physics.optics↗

arXiv2024

Evaluating the performance-deviation of itemKNN in RecBole and LensKit

This study examines the performance of item-based k-Nearest Neighbors (ItemKNN) algorithms in the RecBole and LensKit recommender system libraries. Using four data sets (Anime, Modcloth, ML-100K, and ML-1M), we assess each library's efficiency, accuracy, and scalability, focusing primarily on normalized discounted cumulative gain (nDCG). Our results show that RecBole outperforms LensKit on two of three metrics on the ML-100K data set: it achieved an 18% higher nDCG, 14% higher precision, and 35% lower recall. To ensure a fair comparison, we adjusted LensKit's nDCG calculation to match RecBole's method. This alignment made the performance more comparable, with LensKit achieving an nDCG of 0.2540 and RecBole 0.2674. Differences in similarity matrix calculations were identified as the main cause of performance deviations. After modifying LensKit to retain only the top K similar items, both libraries showed nearly identical nDCG values across all data sets. For instance, both achieved an nDCG of 0.2586 on the ML-1M data set with the same random seed. Initially, LensKit's original implementation only surpassed RecBole in the ModCloth dataset.

cs.LG↗

arXiv2024

Laser driven melt pool resonances through dynamically oscillating energy inputs

Spatially selective melting of metal materials by laser irradiation allows for the precise welding as well as the 3D printing of complex metal parts. However, the simple scanning of a conventional Gaussian beam typically results in a melt track with randomly distributed surface features due to the complex and dynamic behavior of the melt pool. In this study, the implications of utilizing a dynamically oscillating energy input on driving melt track fluctuations is investigated. Specifically, the laser intensity and/or intensity distribution is sinusoidally modulated at different scan speeds, and the effect of modulation frequency on the resulting surface features of the melt track is examined. The formation of periodically oriented surface features indicates an evident frequency coupling between the melt pool and the modulation frequency. Moreover, such a frequency coupling becomes most prominent under a specific modulation frequency, suggesting resonant behavior. The insights provided in this study will enable the development of novel methods, allowing for the control and/or mitigation of inherent fluctuations in the melt pool through laser-driven resonances.

cond-mat.mtrl-sci↗

arXiv2024

Structured Evaluation of Synthetic Tabular Data

Tabular data is common yet typically incomplete, small in volume, and access-restricted due to privacy concerns. Synthetic data generation offers potential solutions. Many metrics exist for evaluating the quality of synthetic tabular data; however, we lack an objective, coherent interpretation of the many metrics. To address this issue, we propose an evaluation framework with a single, mathematical objective that posits that the synthetic data should be drawn from the same distribution as the observed data. Through various structural decomposition of the objective, this framework allows us to reason for the first time the completeness of any set of metrics, as well as unifies existing metrics, including those that stem from fidelity considerations, downstream application, and model-based approaches. Moreover, the framework motivates model-free baselines and a new spectrum of metrics. We evaluate structurally informed synthesizers and synthesizers powered by deep learning. Using our structured framework, we show that synthetic data generators that explicitly represent tabular structure outperform other methods, especially on smaller datasets.

cs.LG↗

arXiv2024

Towards a sensing model using random laser combined with diffuse reflectance spectroscopy

The previous research proves that the random laser emission reflects not only the scattering properties but also the absorption properties. The random laser is therefore considered a potential tool for optical properties sensing. Although the qualitative sensing using the random laser is extensively investigated, a quantitative measurement is still rare. In this study, a generalized mathematical quantitative model using random laser combined with diffuse reflectance spectroscopy is proposed for optical sensing in turbid media. This model describes the gain effect of the active medium and the optical properties effect of the passive medium separately. Rhodamine 6G is used as the active medium. Intralipid and ink are employed to demonstrate the effect of the scattering and absorption, respectively. The peak wavelength shift of the random laser is proved to be an ideal sensing parameter for this sensing model. It is also revealed that the scaling parameters in the sensing model are interrelated and can be simplified to one. With this combined model, the direct sensing of optical properties in diverse turbid media is promising.

physics.optics↗

arXiv2024

Word Importance Explains How Prompts Affect Language Model Outputs

The emergence of large language models (LLMs) has revolutionized numerous applications across industries. However, their "black box" nature often hinders the understanding of how they make specific decisions, raising concerns about their transparency, reliability, and ethical use. This study presents a method to improve the explainability of LLMs by varying individual words in prompts to uncover their statistical impact on the model outputs. This approach, inspired by permutation importance for tabular data, masks each word in the system prompt and evaluates its effect on the outputs based on the available text scores aggregated over multiple user inputs. Unlike classical attention, word importance measures the impact of prompt words on arbitrarily-defined text scores, which enables decomposing the importance of words into the specific measures of interest--including bias, reading level, verbosity, etc. This procedure also enables measuring impact when attention weights are not available. To test the fidelity of this approach, we explore the effect of adding different suffixes to multiple different system prompts and comparing subsequent generations with different large language models. Results show that word importance scores are closely related to the expected suffix importances for multiple scoring functions.

cs.AI↗

arXiv2023

TESS Spots a Super-Puff: The Remarkably Low Density of TOI-1420b

We present the discovery of TOI-1420b, an exceptionally low-density ($ρ= 0.08\pm0.02$ g cm$^{-3}$) transiting planet in a $P = 6.96$ day orbit around a late G dwarf star. Using transit observations from TESS, LCOGT, OPM, Whitin, Wendelstein, OAUV, Ca l'Ou, and KeplerCam along with radial velocity observations from HARPS-N and NEID, we find that the planet has a radius of $R_p$ = 11.9 $\pm$ 0.3 $R_\Earth$ and a mass of $M_p$ = 25.1 $\pm$ 3.8 $M_\Earth$. TOI-1420b is the largest-known planet with a mass less than $50M_\Earth$, indicating that it contains a sizeable envelope of hydrogen and helium. We determine TOI-1420b's envelope mass fraction to be $f_{env} = 82^{+7}_{-6}\%$, suggesting that runaway gas accretion occurred when its core was at most $4-5\times$ the mass of the Earth. TOI-1420b is similar to the planet WASP-107b in mass, radius, density, and orbital period, so a comparison of these two systems may help reveal the origins of close-in low-density planets. With an atmospheric scale height of 1950 km, a transmission spectroscopy metric of 580, and a predicted Rossiter-McLaughlin amplitude of about $17$ m s$^{-1}$, TOI-1420b is an excellent target for future atmospheric and dynamical characterization.

astro-ph.EP↗

arXiv2023

3D Adversarial Augmentations for Robust Out-of-Domain Predictions

Since real-world training datasets cannot properly sample the long tail of the underlying data distribution, corner cases and rare out-of-domain samples can severely hinder the performance of state-of-the-art models. This problem becomes even more severe for dense tasks, such as 3D semantic segmentation, where points of non-standard objects can be confidently associated to the wrong class. In this work, we focus on improving the generalization to out-of-domain data. We achieve this by augmenting the training set with adversarial examples. First, we learn a set of vectors that deform the objects in an adversarial fashion. To prevent the adversarial examples from being too far from the existing data distribution, we preserve their plausibility through a series of constraints, ensuring sensor-awareness and shapes smoothness. Then, we perform adversarial augmentation by applying the learned sample-independent vectors to the available objects when training a model. We conduct extensive experiments across a variety of scenarios on data from KITTI, Waymo, and CrashD for 3D object detection, and on data from SemanticKITTI, Waymo, and nuScenes for 3D semantic segmentation. Despite training on a standard single dataset, our approach substantially improves the robustness and generalization of both 3D object detection and 3D semantic segmentation methods to out-of-domain data.

cs.CV↗

arXiv2023

Segmenting Known Objects and Unseen Unknowns without Prior Knowledge

Panoptic segmentation methods assign a known class to each pixel given in input. Even for state-of-the-art approaches, this inevitably enforces decisions that systematically lead to wrong predictions for objects outside the training categories. However, robustness against out-of-distribution samples and corner cases is crucial in safety-critical settings to avoid dangerous consequences. Since real-world datasets cannot contain enough data points to adequately sample the long tail of the underlying distribution, models must be able to deal with unseen and unknown scenarios as well. Previous methods targeted this by re-identifying already-seen unlabeled objects. In this work, we propose the necessary step to extend segmentation with a new setting which we term holistic segmentation. Holistic segmentation aims to identify and separate objects of unseen, unknown categories into instances without any prior knowledge about them while performing panoptic segmentation of known classes. We tackle this new problem with U3HS, which finds unknowns as highly uncertain regions and clusters their corresponding instance-aware embeddings into individual objects. By doing so, for the first time in panoptic segmentation with unknown objects, our U3HS is trained without unknown categories, reducing assumptions and leaving the settings as unconstrained as in real-life scenarios. Extensive experiments on public data from MS COCO, Cityscapes, and Lost&Found demonstrate the effectiveness of U3HS for this new, challenging, and assumptions-free setting called holistic segmentation. Project page: https://holisticseg.github.io.

cs.CV↗

arXiv2023

Active Learning for Object Detection with Non-Redundant Informative Sampling

Curating an informative and representative dataset is essential for enhancing the performance of 2D object detectors. We present a novel active learning sampling strategy that addresses both the informativeness and diversity of the selections. Our strategy integrates uncertainty and diversity-based selection principles into a joint selection objective by measuring the collective information score of the selected samples. Specifically, our proposed NORIS algorithm quantifies the impact of training with a sample on the informativeness of other similar samples. By exclusively selecting samples that are simultaneously informative and distant from other highly informative samples, we effectively avoid redundancy while maintaining a high level of informativeness. Moreover, instead of utilizing whole image features to calculate distances between samples, we leverage features extracted from detected object regions within images to define object features. This allows us to construct a dataset encompassing diverse object types, shapes, and angles. Extensive experiments on object detection and image classification tasks demonstrate the effectiveness of our strategy over the state-of-the-art baselines. Specifically, our selection strategy achieves a 20% and 30% reduction in labeling costs compared to random selection for PASCAL-VOC and KITTI, respectively.

cs.CV↗

arXiv2023

Monocular 3D Object Detection with LiDAR Guided Semi Supervised Active Learning

We propose a novel semi-supervised active learning (SSAL) framework for monocular 3D object detection with LiDAR guidance (MonoLiG), which leverages all modalities of collected data during model development. We utilize LiDAR to guide the data selection and training of monocular 3D detectors without introducing any overhead in the inference phase. During training, we leverage the LiDAR teacher, monocular student cross-modal framework from semi-supervised learning to distill information from unlabeled data as pseudo-labels. To handle the differences in sensor characteristics, we propose a data noise-based weighting mechanism to reduce the effect of propagating noise from LiDAR modality to monocular. For selecting which samples to label to improve the model performance, we propose a sensor consistency-based selection score that is also coherent with the training objective. Extensive experimental results on KITTI and Waymo datasets verify the effectiveness of our proposed framework. In particular, our selection strategy consistently outperforms state-of-the-art active learning baselines, yielding up to 17% better saving rate in labeling costs. Our training strategy attains the top place in KITTI 3D and birds-eye-view (BEV) monocular object detection official benchmarks by improving the BEV Average Precision (AP) by 2.02.

cs.CV↗

arXiv2023

PG-Schema: Schemas for Property Graphs

Property graphs have reached a high level of maturity, witnessed by multiple robust graph database systems as well as the ongoing ISO standardization effort aiming at creating a new standard Graph Query Language (GQL). Yet, despite documented demand, schema support is limited both in existing systems and in the first version of the GQL Standard. It is anticipated that the second version of the GQL Standard will include a rich DDL. Aiming to inspire the development of GQL and enhance the capabilities of graph database systems, we propose PG-Schema, a simple yet powerful formalism for specifying property graph schemas. It features PG-Types with flexible type definitions supporting multi-inheritance, as well as expressive constraints based on the recently proposed PG-Keys formalism. We provide the formal syntax and semantics of PG-Schema, which meet principled design requirements grounded in contemporary property graph management scenarios, and offer a detailed comparison of its features with those of existing schema languages and graph database systems.

cs.DB↗

arXiv2023

Multi-Task Consistency for Active Learning

Learning-based solutions for vision tasks require a large amount of labeled training data to ensure their performance and reliability. In single-task vision-based settings, inconsistency-based active learning has proven to be effective in selecting informative samples for annotation. However, there is a lack of research exploiting the inconsistency between multiple tasks in multi-task networks. To address this gap, we propose a novel multi-task active learning strategy for two coupled vision tasks: object detection and semantic segmentation. Our approach leverages the inconsistency between them to identify informative samples across both tasks. We propose three constraints that specify how the tasks are coupled and introduce a method for determining the pixels belonging to the object detected by a bounding box, to later quantify the constraints as inconsistency scores. To evaluate the effectiveness of our approach, we establish multiple baselines for multi-task active learning and introduce a new metric, mean Detection Segmentation Quality (mDSQ), tailored for the multi-task active learning comparison that addresses the performance of both tasks. We conduct extensive experiments on the nuImages and A9 datasets, demonstrating that our approach outperforms existing state-of-the-art methods by up to 3.4% mDSQ on nuImages. Our approach achieves 95% of the fully-trained performance using only 67% of the available data, corresponding to 20% fewer labels compared to random selection and 5% fewer labels compared to state-of-the-art selection strategy. Our code will be made publicly available after the review process.

cs.CV↗

arXiv2023

Bridging graph data models: RDF, RDF-star, and property graphs as directed acyclic graphs

Graph database users today face a choice between two technology stacks: the Resource Description Framework (RDF), on one side, is a data model with built-in semantics that was originally developed by the W3C to exchange interconnected data on the Web; on the other side, Labeled Property Graphs (LPGs) are geared towards efficient graph processing and have strong roots in developer and engineering communities. The two models look at graphs from different abstraction layers (triples in RDF vs. edges connecting vertices with inlined properties in LPGs), expose - at least at the surface - distinct features, come with different query languages, and are embedded into their own software ecosystems. In this short paper, we introduce a novel unifying graph data model called Statement Graphs, which combines the traits of both RDF and LPG and achieves interoperability at different levels: it (a) provides the ability to manage RDF and LPG data as a single, interconnected graph, (b) supports querying over the integrated graph using any RDF or LPG query language, while (c) clearing the way for graph stack independent data exchange mechanisms and formats. We formalize our new model as directed acyclic graphs and sketch a system of bidirectional mappings between RDF, LPGs, and Statement Graphs. Our mappings implicitly define read query semantics for RDF and LPGs query languages over the unified data model, thus providing graph users with the flexibility to use the query language of their choice for their graph use cases. As a proof of concept for our ideas, we also present the 1G Playground; an in-memory DBMS built on the concepts of Statement Graphs, which facilitates storage of both RDF and LPG data, and allows for cross-model querying using both SPARQL and Gremlin.

cs.DB↗

arXiv2023

Polarization-Controlled Nonlinear Computer-Generated Holography

Dynamic phase-only beam shaping with a liquid crystal spatial light modulator is a powerful technique for tailoring the beam's intensity profile or wave front. While shaping and controlling the light field is a highly researched topic, dynamic nonlinear beam shaping has hardly been explored so far. One potential reason is that generating the second harmonic is a degenerate process as it mixes two fields at the same frequency. To overcome this problem, we propose the use of type II phase matching as a control mechanism to distinguish the two involved fields. We experimentally demonstrate that arbitrary intensity distributions can be shaped in the frequency-converted field at the same quality as for linear beam shaping and with conversion efficiencies similar to the case without beam shaping. We envision this technique as a milestone towards beam shaping beyond the physical limits of liquid crystal displays, i.e. facilitating dynamic phase-only beam shaping in the ultraviolet spectral range.

physics.optics↗

arXiv2022

Direct laser written aperiodic photonic volume elements for complex light shaping with high efficiency: inverse design and fabrication

Light plays the central role in many applications. The key to unlocking its versatility lies in shaping it into the most appropriate form for the task at hand. Specifically tailored refractive index modifications, directly manufactured inside glass using a short pulsed laser, enable an almost arbitrary control of the light flow. However, the stringent requirements for quantitative knowledge of these modifications, as well as for fabrication precision, have so far prevented the fabrication of light-efficient aperiodic photonic volume elements (APVEs). Here we present a powerful approach to the design and manufacturing of light-efficient APVEs. We optimize application-specific 3D arrangements of hundred thousands of microscopic voxels and manufacture them using femtosecond direct laser writing inside millimeter-sized glass volumes. We experimentally achieve unprecedented diffraction efficiencies up to 80%, which is enabled by precise voxel characterization and adaptive optics during fabrication. We demonstrate APVEs with various functionalities, including a spatial mode converter and combined intensity shaping and wavelength-multiplexing. Our elements can be freely designed and are efficient, compact and robust. Our approach is not limited to borosilicate glass, but is potentially extendable to other substrates, including birefringent and nonlinear materials, giving a preview of even broader functionalities including polarization modulation and dynamic elements.

physics.optics↗

arXiv2022

3D-VField: Adversarial Augmentation of Point Clouds for Domain Generalization in 3D Object Detection

As 3D object detection on point clouds relies on the geometrical relationships between the points, non-standard object shapes can hinder a method's detection capability. However, in safety-critical settings, robustness to out-of-domain and long-tail samples is fundamental to circumvent dangerous issues, such as the misdetection of damaged or rare cars. In this work, we substantially improve the generalization of 3D object detectors to out-of-domain data by deforming point clouds during training. We achieve this with 3D-VField: a novel data augmentation method that plausibly deforms objects via vector fields learned in an adversarial fashion. Our approach constrains 3D points to slide along their sensor view rays while neither adding nor removing any of them. The obtained vectors are transferable, sample-independent and preserve shape and occlusions. Despite training only on a standard dataset, such as KITTI, augmenting with our vector fields significantly improves the generalization to differently shaped objects and scenes. Towards this end, we propose and share CrashD: a synthetic dataset of realistic damaged and rare cars, with a variety of crash scenarios. Extensive experiments on KITTI, Waymo, our CrashD and SUN RGB-D show the generalizability of our techniques to out-of-domain data, different models and sensors, namely LiDAR and ToF cameras, for both indoor and outdoor scenes. Our CrashD dataset is available at https://crashd-cars.github.io.

cs.CV↗

arXiv2021

Graph? Yes! Which one? Help!

Amazon Neptune is a graph database service that supports two graph (meta)models: W3C's Resource Description Framework (RDF) and Labeled Property Graphs (LPG). Customers opt in for one or the other model, and this choice determines which data modeling features can be used, and - perhaps more importantly - which query languages are available to query and manipulate the graph. The choice between the two technology stacks is difficult and requires consideration of data modeling aspects, query language features, their adequacy for current and future use cases, as well as many other factors (including developer preferences). Sometimes we see customers make the wrong choice with no easy way to reverse it later. It is therefore highly desirable that the choice of the query language can be made without consideration of what graph model is chosen, and can be easily revised or complemented at a later point. In this paper, we advocate and explore the idea of a single, unified graph data model that embraces both RDF and LPGs, and naturally supports different graph query languages on top. We investigate obstacles towards unifying the two graph data models, and propose an initial unifying model, dubbed "one graph" ("1G" for short), as the basis for moving forward.

cs.DB↗