Search

The use of reward functions to structure AI learning and decision making is core to the current reinforcement learning paradigm; however, without careful design of reward functions, agents can learn to solve problems in ways that may be considered "undesirable" or "unethical." Without thorough understanding of the incentives a reward function creates, it can be difficult to impose principled yet general control mechanisms over its behavior. In this paper, we study methods for constructing guardrails for AI agents that use reward functions to learn decision making. We introduce a novel approach, which we call strategy masking, to explicitly learn and then suppress undesirable AI agent behavior. We apply our method to study lying in AI agents and show that it can be used to effectively modify agent behavior by suppressing lying post-training without compromising agent ability to perform effectively.

Technological progress related to astronomical observatories such as the recently launched James Webb Space Telescope (JWST) allows searching for signs of life beyond our Solar System, namely in the form of unambiguous biosignature gases in exoplanetary atmospheres. The tentative assignment of a $1-2.4\sigma$ spectral feature observed with JWST in the atmosphere of exoplanet K2-18b to the biosignature gas dimethyl sulfide (DMS; sum formula C$_2$H$_6$S) raised hopes that, although controversial, a second genesis had been found. Terrestrial atmospheric DMS is exclusively stemming from marine biological activity and no natural abiotic source has been identified - neither on Earth nor in space. Therefore, DMS is considered a robust biosignature. Since comets possess a pristine inventory of complex organic molecules of abiotic origin, we have searched high-resolution mass spectra collected at comet 67P/Churyumov-Gerasimenko, target of the European Space Agency's Rosetta mission, for the signatures of DMS. Previous work reported the presence of a C2H6S signal when the comet was near its equinox but distinction of DMS from its structural isomer ethanethiol remained elusive. Here we reassess these and evaluate additional data. Based on differences in the electron ionization induced fragmentation pattern of the two isomers, we show that DMS is significantly better compatible with the observations. Deviations between expected and observed signal intensities for DMS are $<1\sigma$, while for ethanethiol they are $2-4\sigma$. The local abundance of DMS relative to methanol deduced from these data is (0.13$\pm$0.04)%. Our results provide the first evidence for the existence of an abiotic synthetic pathway to DMS in pristine cometary matter and hence motivate more detailed studies of the sulfur chemistry in such matter and its analogs. [...], Comment: 11 pages, 4 figures

We consider a system of two balance laws of Keyfitz-Kranzer type with varying generalized Chaplygin gas, which exhibits negative pressure and is a product of a function of time and the inverse of a power of the density. The Chaplygin gas is a fluid designed to accommodate measurements for the early universe and late-time universal expansion while obeying the pressure-density-time relation. We produce an explanation and description of the non-self-similar Riemann solutions, including the non-classical singular solutions. We also find that due to a direct dependence on time, a change in the regions allowing for combinations of classical and non-classical singular solutions occurs, therefore a Riemann solution can have different solutions over several time intervals. Our findings are confirmed numerically using the local Lax-Friederichs scheme., Comment: 41 pages

The transport of DNA polymers through nanoscale pores is central to many biological processes, from bacterial gene exchange to viral infection. In single-molecule nanopore sensing, the detection of nucleic acid and protein analytes relies on the passage of a long biopolymer through a nanoscale aperture. Understanding the dynamics of polymer translocation through nanopores, especially the relation between ionic current signal and polymer conformations is thus essential for the successful identification of targets. Here, by analyzing ionic current traces of dsDNA translocation, we reveal that features up to now uniquely associated with knots are instead different structural motifs: plectonemes. By combining experiments and simulations, we demonstrate that such plectonemes form because of the solvent flow that induces rotation of the helical DNA fragment in the nanopore, causing torsion propagation outwards from the pore. Molecular dynamic simulations reveal that plectoneme initialization is dominated by the applied torque while the translocation time and size of the plectonemes depend on the coupling of torque and pulling force, a mechanism that might also be relevant for in vivo DNA organization. Experiments with nicked DNA constructs show that the number of plectonemes depends on the rotational constraints of the translocating molecules. Thus, our work introduces plectonemes as essential structural features that must be considered for accurate analysis of polymer transport in the nanopore., Comment: 20 pages, 4 figures

Optical motion capture (mocap) requires accurately reconstructing the human body from retroreflective markers, including pose and shape. In a typical mocap setting, marker labeling is an important but tedious and error-prone step. Previous work has shown that marker labeling can be automated by using a structured template defining specific marker placements, but this places additional recording constraints. We propose to relax these constraints and solve for Unstructured Unlabeled Optical (UUO) mocap. Compared to the typical mocap setting that either labels markers or places them w.r.t a structured layout, markers in UUO mocap can be placed anywhere on the body and even on one specific limb (e.g., right leg for biomechanics research), hence it is of more practical significance. It is also more challenging. To solve UUO mocap, we exploit a monocular video captured by a single RGB camera, which does not require camera calibration. On this video, we run an off-the-shelf method to reconstruct and track a human individual, giving strong visual priors of human body pose and shape. With both the video and UUO markers, we propose an optimization pipeline towards marker identification, marker labeling, human pose estimation, and human body reconstruction. Our technical novelties include multiple hypothesis testing to optimize global orientation, and marker localization and marker-part matching to better optimize for body surface. We conduct extensive experiments to quantitatively compare our method against state-of-the-art approaches, including marker-only mocap and video-only human body/shape reconstruction. Experiments demonstrate that our method resoundingly outperforms existing methods on three established benchmark datasets for both full-body and partial-body reconstruction.

We generalize 2-Wasserstein dependence coefficients to measure dependence between a finite number of random vectors. This generalization includes theoretical properties, and in particular focuses on an interpretation of maximal dependence and an asymptotic normality result for a proposed semi-parametric estimator under a Gaussian copula assumption. In addition, we discuss general axioms for dependence measures between multiple random vectors, other plausible normalizations, and various examples. Afterwards, we look into plug-in estimators based on penalized empirical covariance matrices in order to deal with high dimensionality issues and take possible marginal independencies into account by inducing (block) sparsity. The latter ideas are investigated via a simulation study, considering other dependence coefficients as well. We illustrate the use of the developed methods in two real data applications.

We investigate how to use AI-based analytics to support design education. The analytics at hand measure multiscale design, that is, students' use of space and scale to visually and conceptually organize their design work. With the goal of making the analytics intelligible to instructors, we developed a research artifact integrating a design analytics dashboard with design instances, and the design environment that students use to create them. We theorize about how Suchman's notion of mutual intelligibility requires contextualized investigation of AI in order to develop findings about how analytics work for people. We studied the research artifact in 5 situated course contexts, in 3 departments. A total of 236 students used the multiscale design environment. The 9 instructors who taught those students experienced the analytics via the new research artifact. We derive findings from a qualitative analysis of interviews with instructors regarding their experiences. Instructors reflected on how the analytics and their presentation in the dashboard have the potential to affect design education. We develop research implications addressing: (1) how indexing design analytics in the dashboard to actual design work instances helps design instructors reflect on what they mean and, more broadly, is a technique for how AI-based design analytics can support instructors' assessment and feedback experiences in situated course contexts; and (2) how multiscale design analytics, in particular, have the potential to support design education. By indexing, we mean linking which provides context, here connecting the numbers of the analytics with visually annotated design work instances., Comment: 22 pages, 4 figures, Submitted to ACM DIS

Recent experiments have probed the relative likelihoods of trajectories in stochastic systems by observing survival probabilities within a tube of radius $R$ in spacetime. We measure such probabilities here for a colloidal particle in a corrugated channel, corresponding to a bistable potential with state-dependent diffusivity. In contrast to previous findings for state-independent noise, we find that the most probable pathway changes qualitatively as the tube radius $R$ is altered. We explain this by computing the survival probabilities predicted by overdamped Langevin dynamics. At high enough resolution (small enough $R$), survival probabilities depend solely on diffusivity variations, independent of deterministic forces; finite $R$ corrections yield a generalization of the Onsager-Machlup action. As corollary, ratios of survival probabilities are singular as $R \to 0$, but become regular, and described by the classical Onsager-Machlup action, only in the special case of state-independent noise., Comment: 6 pages, 3 figures. Note also the accompanying manuscript arXiv:2009.04250

Neural implicit fields, such as the neural signed distance field (SDF) of a shape, have emerged as a powerful representation for many applications, e.g., encoding a 3D shape and performing collision detection. Typically, implicit fields are encoded by Multi-layer Perceptrons (MLP) with positional encoding (PE) to capture high-frequency geometric details. However, a notable side effect of such PE-equipped MLPs is the noisy artifacts present in the learned implicit fields. While increasing the sampling rate could in general mitigate these artifacts, in this paper we aim to explain this adverse phenomenon through the lens of Fourier analysis. We devise a tool to determine the appropriate sampling rate for learning an accurate neural implicit field without undesirable side effects. Specifically, we propose a simple yet effective method to estimate the intrinsic frequency of a given network with randomized weights based on the Fourier analysis of the network's responses. It is observed that a PE-equipped MLP has an intrinsic frequency much higher than the highest frequency component in the PE layer. Sampling against this intrinsic frequency following the Nyquist-Sannon sampling theorem allows us to determine an appropriate training sampling rate. We empirically show in the setting of SDF fitting that this recommended sampling rate is sufficient to secure accurate fitting results, while further increasing the sampling rate would not further noticeably reduce the fitting error. Training PE-equipped MLPs simply with our sampling strategy leads to performances superior to the existing methods.

ESA's Rosetta spacecraft at comet 67P/Churyumov-Gerasimenko (67P) was the first mission that accompanied a comet over a substantial fraction of its orbit. On board was the ROSINA mass spectrometer suite to measure the local densities of the volatile species sublimating from the ices inside the comet's nucleus. Understanding the nature of these ices was a key goal of Rosetta. We analyzed the primary cometary molecules at 67P, namely H$_2$O and CO$_2$, together with a suite of minor species for almost the entire mission. Our investigation reveals that the local abundances of highly volatile species, such as CH$_4$ and CO, are reproduced by a linear combination of both H$_2$O and CO$_2$ densities. These findings bear similarities to laboratory-based temperature programmed desorption experiments of amorphous ices and imply that highly volatile species are trapped in H$_2$O and CO$_2$ ices. Our results do not show the presence of ices dominated by these highly volatile molecules. Most likely, they were lost due to thermal processing of 67P's interior prior to its deflection to the inner solar system. Deviations in the proportions co-released with H$_2$O and CO$_2$ can only be observed before the inbound equinox, when the comet was still far from the sun and the abundance of highly volatile molecules associated with CO$_2$ outgassing were lower. The corresponding CO$_2$ is likely seasonal frost, which sublimated and lost its trapped highly volatile species before re-freezing during the previous apparition. CO, on the other hand, was elevated during the same time and requires further investigation., Comment: This is a pre-copyedited, author-produced PDF of an article accepted for publication in Monthly Notices of the Royal Astronomical Society following peer review. The version of record is available online at: https://doi.org/10.1093/mnras/stad3005

The recent surge of utilizing deep neural networks for geometric processing and shape modeling has opened up exciting avenues. However, there is a conspicuous lack of research efforts on using powerful neural representations to extend the capabilities of parametric surfaces, which are the prevalent surface representations in product design, CAD/CAM, and computer animation. We present Neural Parametric Surfaces, the first piecewise neural surface representation that allows coarse patch layouts of arbitrary $n$-sided surface patches to model complex surface geometries with high precision, offering greater flexibility over traditional parametric surfaces. By construction, this new surface representation guarantees $G^0$ continuity between adjacent patches and empirically achieves $G^1$ continuity, which cannot be attained by existing neural patch-based methods. The key ingredient of our neural parametric surface is a learnable feature complex $\mathcal{C}$ that is embedded in a high-dimensional space $\mathbb{R}^D$ and topologically equivalent to the patch layout of the surface; each face cell of the complex is defined by interpolating feature vectors at its vertices. The learned feature complex is mapped by an MLP-encoded function $f:\mathcal{C} \rightarrow \mathcal{S}$ to produce the neural parametric surface $\mathcal{S}$. We present a surface fitting algorithm that optimizes the feature complex $\mathcal{C}$ and trains the neural mapping $f$ to reconstruct given target shapes with high accuracy. We further show that the proposed representation along with a compact-size neural net can learn a plausible shape space from a shape collection, which can be used for shape interpolation or shape completion from noisy and incomplete input data. Extensive experiments show that neural parametric surfaces offer greater modeling capabilities than traditional parametric surfaces., Comment: 17 pages, 16 figures

We propose a method, named DualMesh-UDF, to extract a surface from unsigned distance functions (UDFs), encoded by neural networks, or neural UDFs. Neural UDFs are becoming increasingly popular for surface representation because of their versatility in presenting surfaces with arbitrary topologies, as opposed to the signed distance function that is limited to representing a closed surface. However, the applications of neural UDFs are hindered by the notorious difficulty in extracting the target surfaces they represent. Recent methods for surface extraction from a neural UDF suffer from significant geometric errors or topological artifacts due to two main difficulties: (1) A UDF does not exhibit sign changes; and (2) A neural UDF typically has substantial approximation errors. DualMesh-UDF addresses these two difficulties. Specifically, given a neural UDF encoding a target surface $\bar{S}$ to be recovered, we first estimate the tangent planes of $\bar{S}$ at a set of sample points close to $\bar{S}$. Next, we organize these sample points into local clusters, and for each local cluster, solve a linear least squares problem to determine a final surface point. These surface points are then connected to create the output mesh surface, which approximates the target surface. The robust estimation of the tangent planes of the target surface and the subsequent minimization problem constitute our core strategy, which contributes to the favorable performance of DualMesh-UDF over other competing methods. To efficiently implement this strategy, we employ an adaptive Octree. Within this framework, we estimate the location of a surface point in each of the octree cells identified as containing part of the target surface. Extensive experiments show that our method outperforms existing methods in terms of surface reconstruction quality while maintaining comparable computational efficiency., Comment: ICCV 2023

Neural implicit representations are known to be more compact for depicting 3D shapes than traditional discrete representations. However, the neural representations tend to round sharp corners or edges and struggle to represent surfaces with open boundaries. Moreover, they are slow to train. We present a unified neural implicit representation, called Patch-Grid, that fits to complex shapes efficiently, preserves sharp features, and effectively models surfaces with open boundaries and thin geometric features. Our superior efficiency comes from embedding each surface patch into a local latent volume and decoding it using a shared MLP decoder, which is pretrained on various local surface geometries. With this pretrained decoder fixed, fitting novel shapes and local shape updates can be done efficiently. The faithful preservation of sharp features is enabled by adopting a novel merge grid to perform local constructive solid geometry (CSG) combinations of surface patches in the cells of an adaptive Octree, yielding better robustness than using a global CSG construction as proposed in the literature. Experiments show that our Patch-Grid method faithfully captures shapes with complex sharp features, open boundaries and thin structures, and outperforms existing learning-based methods in both efficiency and quality for surface fitting and local shape updates.

The puzzling complexity of terrestrial biomolecules is driving the search for complex organic molecules in the Interstellar Medium (ISM) and serves as a motivation for many in situ studies of reservoirs of extraterrestrial organics from meteorites and interplanetary dust particles (IDPs) to comets and asteroids. Comet 67P/Churyumov-Gerasimenko (67P) -- the best-studied comet to date -- has been visited and accompanied for two years by the European Space Agency's Rosetta spacecraft. Around 67P's perihelion and under dusty conditions, the high-resolution mass spectrometer on board provided a spectacular glimpse into this comet's chemical complexity. For this work, we analyzed in unprecedented detail the O-bearing organic volatiles. In a comparison of 67P's inventory to molecules detected in the ISM, in other comets, and in Soluble Organic Matter (SOM) extracted from the Murchison meteorite, we also highlight the (pre)biotic relevance of different chemical groups of species. We report first evidence for abundant extraterrestrial O-bearing heterocycles (with abundances relative to methanol often on the order of 10% with a relative error margin of 30-50%) and various representatives of other molecule classes such as carboxylic acids and esters, aldehydes, ketones, and alcohols. Like with the pure hydrocarbons, some hydrogenated forms seem to be dominant over their dehydrogenated counterparts. An interesting example is tetrahydrofuran (THF) as it might be a more promising candidate for searches in the ISM than the long-sought furan itself. Our findings not only support and guide future efforts to investigate the origins of chemical complexity in space, but also they strongly encourage studies of, e.g., the ratios of unbranched vs. branched and hydrogenated vs. dehydrogenated species in astrophysical ice analogs in the laboratory as well as by modeling., Comment: Reproduced with permission from Astronomy & Astrophysics, copyright ESO

This article proposes copula-based dependence quantification between multiple groups of random variables of possibly different sizes via the family of $Phi$-divergences. An axiomatic framework for this purpose is provided, after which we focus on the absolutely continuous setting assuming copula densities exist. We consider parametric and semi-parametric frameworks, discuss estimation procedures, and report on asymptotic properties of the proposed estimators. In particular, we first concentrate on a Gaussian copula approach yielding explicit and attractive dependence coefficients for specific choices of $Phi$, which are more amenable for estimation. Next, general parametric copula families are considered, with special attention to nested Archimedean copulas, being a natural choice for dependence modelling of random vectors. The results are illustrated by means of examples. Simulations and a real-world application on financial data are provided as well.

The Lunar Orbital Platform-Gateway (LOP-Gateway, or simply Gateway) is a crewed platform that will be assembled and operated in the vicinity of the Moon by NASA and international partner organizations, including ESA, starting from the mid-2020s. It will offer new opportunities for fundamental and applied scientific research. The Moon is a unique location to study the deep space plasma environment. Moreover, the lunar surface and the surface-bounded exosphere are interacting with this environment, constituting a complex multi-scale interacting system. This paper examines the opportunities provided by externally mounted payloads on the Gateway in the field of space plasma physics, heliophysics and space weather, but also examines the impact of the space environment on an inhabited platform in the vicinity of the Moon. It then presents the conceptual design of a model payload, required to perform these space plasma measurements and observations. It results that the Gateway is very well-suited for space plasma physics research. It allows a series of scientific objectives with a multidisciplinary dimension to be addressed.