Search

This study explores the computational mechanisms underlying the resolution of cognitive dissonances. We focus on scenarios in which an observation violates the expected relationship between objects. For instance, an agent expects object A to be smaller than B in some feature space but observes the opposite. One solution is to adjust the expected relationship according to the new observation and change the expectation to A being larger than B. An alternative solution would be to adapt the representation of A and B in the feature space such that in the new representation, the relationship that A is smaller than B is maintained. While both pathways resolve the dissonance, they generalize differently to different tasks. Using Artificial Neural Networks (ANNs) capable of relational learning, we demonstrate the existence of these two pathways and show that the chosen pathway depends on the dissonance magnitude. Large dissonances alter the representation of the objects, while small dissonances lead to adjustments in the expected relationships. We show that this effect arises from the inherently different learning dynamics of relationships and representations and study the implications.

The nature of abstract reasoning is a matter of debate. Modern artificial neural network (ANN) models, like large language models, demonstrate impressive success when tested on abstract reasoning problems. However, it has been argued that their success reflects some form of memorization of similar problems (data contamination) rather than a general-purpose abstract reasoning capability. This concern is supported by evidence of brittleness, and the requirement of extensive training. In our study, we explored whether abstract reasoning can be achieved using the toolbox of ANNs, without prior training. Specifically, we studied an ANN model in which the weights of a naive network are optimized during the solution of the problem, using the problem data itself, rather than any prior knowledge. We tested this modeling approach on visual reasoning problems and found that it performs relatively well. Crucially, this success does not rely on memorization of similar problems. We further suggest an explanation of how it works. Finally, as problem solving is performed by changing the ANN weights, we explored the connection between problem solving and the accumulation of knowledge in the ANNs.

In recent years, Large Language Models (LLM) have demonstrated impressive capabilities in the field of natural language processing (NLP). This paper explores the application of LLMs in negotiation transcript analysis by the Vanderbilt AI Negotiation Lab. Starting in September 2022, we applied multiple strategies using LLMs from zero shot learning to fine tuning models to in-context learning). The final strategy we developed is explained, along with how to access and use the model. This study provides a sense of both the opportunities and roadblocks for the implementation of LLMs in real life applications and offers a model for how LLMs can be applied to coding in other fields.

The in vitro scratch assay is a widely used assay in cell biology to assess the rate of wound closure related to a variety of therapeutic interventions. While manual measurement is subjective and vulnerable to intra- and interobserver variability, computer-based tools are theoretically objective, but in practice often contain parameters which are manually adjusted (individually per image or data set) and thereby provide a source for subjectivity. Modern deep learning approaches typically require large annotated training data which complicates instant applicability. In this paper, we make use of the segment anything model, a deep foundation model based on interactive point-prompts, which enables class-agnostic segmentation without tuning the network's parameters based on domain specific training data. The proposed method clearly outperformed a semi-objective baseline method that required manual inspection and, if necessary, adjustment of parameters per image. Even though the point prompts of the proposed approach are theoretically also a source for subjectivity, results attested very low intra- and interobserver variability, even compared to manual segmentation of domain experts.

The origin of a compact millimeter (mm, 100-250 GHz) emission in radio-quiet active galactic nuclei (RQ AGN) remains debated. Recent studies propose a connection with self-absorbed synchrotron emission from the accretion disk X-ray corona. We present the first joint ALMA ($\sim$100 GHz) and X-ray (NICER/XMM-Newton/Swift; 2-10 keV) observations of the unobscured RQ AGN, IC 4329A ($z = 0.016$). The time-averaged mm-to-X-ray flux ratio aligns with recently established trends for larger samples (Kawamuro et al. 2022, Ricci et al. 2023), but with a tighter scatter ($\sim$0.1 dex) compared to previous studies. However, there is no significant correlation on timescales of less than 20 days. The compact mm emission exhibits a spectral index of $-0.23 \pm 0.18$, remains unresolved with a 13 pc upper limit, and shows no jet signatures. Notably, the mm flux density varies significantly (factor of 3) within 4 days, exceeding the contemporaneous X-ray variability (37% vs. 18%) and showing the largest mm variations ever detected in RQ AGN over daily timescales. The high amplitude variability rules out scenarios of heated dust and thermal free-free emission, pointing toward a synchrotron origin for the mm radiation in a source of $\sim$1 light day size. While the exact source is not yet certain, an X-ray corona scenario emerges as the most plausible compared to a scaled-down jet or outflow-driven shocks.}, Comment: Submitted to A&A

An accretion disk formed around a supermassive black hole (SMBH) after it disrupts a star is expected to be initially misaligned with respect to the black hole's equatorial plane. This misalignment induces relativistic torques (the Lense-Thirring effect) on the disk, causing the disk to precess at early times, while at late times the disk aligns with the black hole and precession terminates. Here, using high-cadence X-ray monitoring observations of a TDE, we report the discovery of strong, quasi-periodic X-ray flux and temperature modulations from a TDE. These X-ray modulations are separated by 17.0$^{+1.2}_{-2.4}$ days and persist for roughly 130 days during the early phase of the TDE. Lense-Thirring precession of the accretion flow can produce this X-ray variability, but other physical mechanisms, such as the radiation-pressure instability, cannot be ruled out. Assuming typical TDE parameters, i.e., a solar-like star with the resulting disk extending at-most to so-called circularization radius, and that the disk precesses as a rigid body, we constrain the disrupting black hole's dimensionless spin parameter to be 0.05<|a|<0.5., Comment: Accepted for publication in Nature

We present the results of a joint {\it XMM-Newton} and {\it NuSTAR} campaign on the active galactic nucleus (AGN) IC 4329A, consisting of 9 $\times$ 20 ks {\it XMM-Newton} observations, and 5 $\times$ 20 ks {\it NuSTAR} observations within nine days, performed in August 2021. Within each observation, the AGN is not very variable, with the fractional variability never exceeding 5%. Flux variations are observed between the different observations, on timescales of days, with a 30% ratio between the minimum and the maximum 2-10 keV flux. These variations follow the softer-when-brighter behavior typically observed in AGN. In all observations, a soft excess is clearly present. Consistently with previous observations, the X-ray spectra of the source exhibit a cut-off energy between 140 and 250 keV, constant within the error in the different observations. We detected a prominent component of the $6.4$\,keV Fe~K$\alpha$ line consistent with being constant during the monitoring, consisting of an unresolved narrow core and a broader component likely originating in the inner accredion disc. We find that the reflection component is weak ($R_{\rm max}=0.009\pm0.002$) and most likely originating in distant neutral medium. We also found the presence of a warm absorber component together with an ultra-fast outflow. Looking at their energetic, these outflows have enough mechanical power to exercise a significant feedback impact on the AGN surrounding environment., Comment: Accepted for publication in Astronomy & Astrophysics

We present the first results from a 100-day Swift, NICER and ground-based X-ray/UV/optical reverberation mapping campaign of the Narrow-Line Seyfert 1 Mrk 335, when it was in an unprecedented low X-ray flux state. Despite dramatic suppression of the X-ray variability, we still observe UV/optical lags as expected from disk reverberation. Moreover, the UV/optical lags are consistent with archival observations when the X-ray luminosity was >10 times higher. Interestingly, both low- and high-flux states reveal UV/optical lags that are 6-11 times longer than expected from a thin disk. These long lags are often interpreted as due to contamination from the broad line region, however the u band excess lag (containing the Balmer jump from the diffuse continuum) is less prevalent than in other AGN. The Swift campaign showed a low X-ray-to-optical correlation (similar to previous campaigns), but NICER and ground-based monitoring continued for another two weeks, during which the optical rose to the highest level of the campaign, followed ~10 days later by a sharp rise in X-rays. While the low X-ray countrate and relatively large systematic uncertainties in the NICER background make this measurement challenging, if the optical does lead X-rays in this flare, this indicates a departure from the zeroth-order reprocessing picture. If the optical flare is due to an increase in mass accretion rate, this occurs on much shorter than the viscous timescale. Alternatively, the optical could be responding to an intrinsic rise in X-rays that is initially hidden from our line-of-sight., Comment: Accepted for publication in the Astrophysical Journal. 15 pages, 8 figures, 3 tables

We present XSLIDE (X-Ray Spectral Line IDentifier and Explorer), a graphical user interface that has been designed as a quick-look tool for the upcoming X-Ray Imaging and Spectroscopy Mission (XRISM). XSLIDE is a simple and user-friendly application that allows for the interactive plotting of spectra from XRISM's Resolve instrument without requiring the selection of models for forward-fitting. XSLIDE performs common tasks such as rebinning, continuum fitting, automatically detecting lines, assigning detected lines to known atomic transitions, spectral diagnostics, and more. It is expected that XSLIDE will allow XRISM's scientific investigators to rapidly examine many spectra to find those which contain spectral lines of particular interest, and it will also allow astronomers from outside the field of high-resolution X-ray spectroscopy to easily interact with XRISM data., Comment: 10 pages, 6 figures, 1 ancillary video

XRISM (X-Ray Imaging and Spectroscopy Mission), with the Resolve high-resolution spectrometer and the Xtend wide-field imager on-board, is designed to build on the successes of the abbreviated Hitomi mission to address outstanding astrophysical questions using high resolution X-ray spectroscopy. In preparation for launch, the XRISM Science Data Center (SDC) is constructing and testing an integrated and automated system for data transfer and processing based upon the Hitomi framework, introducing improvements informed by previous experience and internal collaboration. The XRISM pipeline ingests FITS files transferred from Japan that contain data converted from spacecraft telemetry, processes (calibrates and screens) the data, creates data products, and transfers data and metadata used to populate data archives in the U.S. and Japan. Improvement and rigorous testing of the system are conducted from the single-task level through fully-integrated levels. We provide an overview of the XRISM pipeline system, with a focus on the data processing, and how new and improved documentation and testing are creating accessible and effective software tools for future XRISM data., Comment: 13 pages, 7 figures, SPIE Conference: Space Telescopes and Instrumentation 2022: Ultraviolet to Gamma Ray

1ES 1927+654 is a paradigm-defying AGN and one of the most peculiar X-ray nuclear transients. In early 2018, this well-known AGN underwent a changing-look event, in which broad optical emission lines appeared and the optical flux increased. Yet, by July 2018, the X-ray flux had dropped by over two orders of magnitude, indicating a dramatic change to the inner accretion flow. With three years of observations with NICER, XMM-Newton, and NuSTAR, we present the X-ray evolution of 1ES 1927+654, which can be broken into three phases-(1) an early super-Eddington phase with rapid variability in X-ray luminosity and spectral parameters, (2) a stable super-Eddington phase at the peak X-ray luminosity, and (3) a steady decline back to the pre-outburst luminosity and spectral parameters. For the first time, we witnessed the formation of the X-ray corona, as the X-ray spectrum transitioned from thermally-dominated to primarily Comptonized. We also track the evolution of the prominent, broad 1 keV feature in the early X-ray spectra and show that this feature can be modeled with blueshifted reflection (z = -0.33) from a single-temperature blackbody irradiating spectrum using xillverTDE, a new flavor of the xillver models. Thus, we propose that the 1 keV feature could arise from reflected emission off the base of an optically thick outflow from a geometrically thick, super-Eddington inner accretion flow, connecting the inner accretion flow with outflows launched during extreme accretion events (e.g. tidal disruption events). Lastly, we compare 1ES 1927+654 to other nuclear transients and discuss applications of xillverTDE to super-Eddington accretors., Comment: 25 pages, 10 figures, accepted for publication in ApJ

In this paper, we aimed to help bridge the gap between human fluid intelligence - the ability to solve novel tasks without prior training - and the performance of deep neural networks, which typically require extensive prior training. An essential cognitive component for solving intelligence tests, which in humans are used to measure fluid intelligence, is the ability to identify regularities in sequences. This motivated us to construct a benchmark task, which we term \textit{sequence consistency evaluation} (SCE), whose solution requires the ability to identify regularities in sequences. Given the proven capabilities of deep networks, their ability to solve such tasks after extensive training is expected. Surprisingly, however, we show that naive (randomly initialized) deep learning models that are trained on a \textit{single} SCE with a \textit{single} optimization step can still solve non-trivial versions of the task relatively well. We extend our findings to solve, without any prior training, real-world anomaly detection tasks in the visual and auditory modalities. These results demonstrate the fluid-intelligent computational capabilities of deep networks. We discuss the implications of our work for constructing fluid-intelligent machines.

Video prediction models often combine three components: an encoder from pixel space to a small latent space, a latent space prediction model, and a generative model back to pixel space. However, the large and unpredictable pixel space makes training such models difficult, requiring many training examples. We argue that finding a predictive latent variable and using it to evaluate the consistency of a future image enables data-efficient predictions because it precludes the necessity of a generative model training. To demonstrate it, we created sequence completion intelligence tests in which the task is to identify a predictably changing feature in a sequence of images and use this prediction to select the subsequent image. We show that a one-dimensional Markov Contrastive Predictive Coding (M-CPC_1D) model solves these tests efficiently, with only five examples. Finally, we demonstrate the usefulness of M-CPC_1D in solving two tasks without prior training: anomaly detection and stochastic movement video prediction.