Search

We report the detection of stimulated hydrogen radio recombination line (RRL) emission from ionized gas in a $z=0.89$ galaxy using 580--1670 MHz observations from the MeerKAT Absorption Line Survey (MALS). The RRL emission originates in a galaxy that intercepts and strongly lenses the radio blazar PKS 1830-211 ($z=2.5$). This is the second detection of RRLs outside of the local universe and the first clearly associated with hydrogen. We detect effective H144$\alpha$ (and H163$\alpha$) transitions at observed frequencies of 1156 (798) MHz by stacking 17 (27) RRLs with 21$\sigma$ (14$\sigma$) significance. The RRL emission contains two main velocity components and is coincident in velocity with HI 21 cm and OH 18 cm absorption. We use the RRL spectral line energy distribution and a Bayesian analysis to constrain the density ($n_e$) and the volume-averaged pathlength ($\ell$) of the ionized gas. We determine $\log( n_e ) = 2.0_{-0.7}^{+1.0}$ cm$^{-3}$ and $\log( \ell ) = -0.7_{-1.1}^{+1.1}$ pc towards the north east (NE) lensed image, likely tracing the diffuse thermal phase of the ionized ISM in a thin disk. Towards the south west (SW) lensed image, we determine $\log( n_e ) = 3.2_{-1.0}^{+0.4}$ cm$^{-3}$ and $\log( \ell ) = -2.7_{-0.2}^{+1.8}$ pc, tracing gas that is more reminiscent of H II regions. We estimate a star formation (surface density) rate of $\Sigma_{\mathrm{SFR}} \sim 0.6$ M$_{\odot}$ yr$^{-1}$ kpc$^{-2}$ or SFR $\sim 50$ M$_{\odot}$ yr$^{-1}$, consistent with a star-forming main sequence galaxy of $M_{\star} \sim 10^{11}$ M$_{\odot}$. The discovery presented here opens up the possibility of studying ionized gas at high redshifts using RRL observations from current and future (e.g., SKA and ngVLA) radio facilities., Comment: 17 pages, 10 figures, 2 tables, accepted in ApJ

In this era of fast and large-scale opinion formation, a mathematical understanding of opinion evolution, a.k.a. opinion dynamics, acquires importance. Linear graph-based dynamics and bounded confidence dynamics are the two popular models for opinion dynamics in social networks. Stochastic bounded confidence (SBC) opinion dynamics was proposed as a general framework that incorporates both these dynamics as special cases and also captures the inherent stochasticity and noise (errors) in real-life social exchanges. Although SBC dynamics is quite general and realistic, its analysis is more challenging. This is because SBC dynamics is nonlinear and stochastic, and belongs to the class of Markov processes that have asymptotically zero drift and unbounded jumps. The asymptotic behavior of SBC dynamics was characterized in prior works. However, they do not shed light on its finite-time behavior, which is often of interest in practice. We take a stride in this direction by analyzing the finite-time behavior of a two-agent system and a bistar graph, which are crucial to the understanding of general multi-agent dynamics. In particular, we show that the opinion difference between the two agents is well-concentrated around zero under the conditions that lead to asymptotic stability of the SBC dynamics., Comment: A preliminary version of this paper appeared in the proceedings of COMmunication Systems & NETworkS (COMSNETS) 2022. arXiv admin note: substantial text overlap with arXiv:2112.04373

Scoring the factuality of a generated summary involves measuring the degree to which a target text contains factual information using the input document as support. Given the similarities in the problem formulation, previous work has shown that Natural Language Inference models can be effectively repurposed to perform this task. As these models are trained to score entailment at a sentence level, several recent studies have shown that decomposing either the input document or the summary into sentences helps with factuality scoring. But is fine-grained decomposition always a winning strategy? In this paper we systematically compare different granularities of decomposition -- from document to sub-sentence level, and we show that the answer is no. Our results show that incorporating additional context can yield improvement, but that this does not necessarily apply to all datasets. We also show that small changes to previously proposed entailment-based scoring methods can result in better performance, highlighting the need for caution in model and methodology selection for downstream tasks., Comment: Generation, Evaluation & Metrics (GEM) Workshop 2022

We consider maximizing the long-term average reward in a single server queue, where the reward obtained for a job is a non-increasing function of its sojourn time. The motivation behind this work comes from multiple applications, including quantum information processing and multimedia streaming. We introduce a new service discipline, shortest predicted sojourn time (SPST), which, in simulations, performs better than well-known disciplines. We also present some limited analytical guarantees for this highly intricate problem., Comment: 8 pages, 6 figures

We present deep L-Band observations of the equatorial field centered on the z=6.3 SDSS QSO, reaching a 1 sigma sensitivity of ~2.5 uJy at the center of the field. We extracted a catalog of 1489 radio sources down to a flux density of ~12.5 uJy (5 sigma) over a field of view of ~ 30' diameter. We derived the source counts accounting for catalog reliability and completeness, and compared them with others available in the literature. Our source counts are among the deepest available so far, and, overall, are consistent with recent counts' determinations and models. We detected for the first time in the radio band the SDSS J1030+0524 QSO (26 +/- 5 uJy). We derived its optical radio loudness R_O = 0.62 +/- 0.12, which makes it the most radio quiet AGN at z >~ 6 discovered so far and detected at radio wavelengths. We unveiled extended diffuse radio emission associated with the lobes of a bright FRII radio galaxy located close to the center of the J1030 field, which is likely to become the future BCG of a protocluster at z=1.7. The lobes' complex morphology, coupled with the presence of X-ray diffuse emission detected around the FRII galaxy lobes, may point toward an interaction between the radio jets and the external medium. We also investigated the relation between radio and X-ray luminosity for a sample of 243 X-ray-selected objects obtained from 500 ks Chandra observations of the same field, and spanning a wide redshift range (0 ~< z ~< 3). Focused on sources with a spectroscopic redshift and classification, we found that sources hosted by ETG and AGN follow Log(L_R)/Log(L_X) linear correlations with slopes of ~0.6 and ~0.8, respectively. This is interpreted as a likely signature of different efficiency in the accretion process. Finally, we found that most of these sources (>~87%) show a radio-to-X-ray radio loudness R_X < -3.5, classifying these objects as radio quiet., Comment: 18 pages, 15 figures, 5 table. Accepted for publication in A&A

CASA, the Common Astronomy Software Applications, is the primary data processing software for the Atacama Large Millimeter/submillimeter Array (ALMA) and the Karl G. Jansky Very Large Array (VLA), and is frequently used also for other radio telescopes. The CASA software can handle data from single-dish, aperture-synthesis, and Very Long Baseline Interferometery (VLBI) telescopes. One of its core functionalities is to support the calibration and imaging pipelines for ALMA, VLA, VLA Sky Survey (VLASS), and the Nobeyama 45m telescope. This paper presents a high-level overview of the basic structure of the CASA software, as well as procedures for calibrating and imaging astronomical radio data in CASA. CASA is being developed by an international consortium of scientists and software engineers based at the National Radio Astronomical Observatory (NRAO), the European Southern Observatory (ESO), the National Astronomical Observatory of Japan (NAOJ), and the Joint Institute for VLBI European Research Infrastructure Consortium (JIV-ERIC), under the guidance of NRAO., Comment: Accepted for publication in PASP (20 pages, 4 figures). Joint publication with CASA-VLBI paper

We consider critical eigenstates in a two dimensional quasicrystal and their evolution as a function of disorder. By exact diagonalization of finite size systems we show that the evolution of properties of a typical wave-function is non-monotonic. That is, disorder leads to states delocalizing, until a certain crossover disorder strength is attained, after which they start to localize. Although this non-monotonic behavior is only present in finite-size systems and vanishes in the thermodynamic limit, the crossover disorder strength decreases logarithmically slowly with system size, and is quite large even for very large approximants. The non-monotonic evolution of spatial properties of eigenstates can be observed in the anomalous dimensions of the wave-function amplitudes, in their multifractal spectra, and in their dynamical properties. We compute the two-point correlation functions of wave-function amplitudes and show that these follow power laws in distance and energy, consistent with the idea that wave-functions retain their multifractal structure on a scale which depends on disorder strength. Dynamical properties are studied as a function of disorder. We find that the diffusion exponents do not reflect the non-monotonic wave-function evolution. Instead, they are essentially independent of disorder until disorder increases beyond the crossover value, after which they decrease rapidly, until the strong localization regime is reached. The differences between our results and earlier studies on geometrically disordered ``phason-flip'' models lead us to propose that the two models are in different universality classes. We conclude by discussing some implications of our results for transport and a proposal for a Mott hopping mechanism between power law localized wave-functions, in moderately disordered quasicrystals., Comment: New figures and references added (23 pages, 13 figures)

We consider a two-dimensional generalization of the Su-Schrieffer-Heeger model which is known to possess a non-trivial topological band structure. For this model, which is characterized by a single parameter, the hopping ratio $0 \leq r\leq 1$, the inhomogeneous superconducting phases induced by an attractive $U$ Hubbard interaction are studied using mean field theory. We show, analytically and by numerical diagonalization, that in lattices with open boundaries, phases with enhanced superconducting order on the corners or the edges can appear, depending on the filling. For finite samples at half filling, the corner site superconducting transition temperature can be much larger than that of the bulk. A novel proximity effect thus arises for $T_{c,bulk} < T

Component-based synthesis seeks to build programs using the APIs provided by a set of libraries. Oftentimes, these APIs have effects, which make it challenging to reason about the correctness of potential synthesis candidates. This is because changes to global state made by effectful library procedures affect how they may be composed together, yielding an intractably large search space that can confound typical enumerative synthesis techniques. If the nature of these effects are exposed as part of their specification, however, deductive synthesis approaches can be used to help guide the search for components. In this paper, we present a new specification-guided synthesis procedure that uses Hoare-style pre- and post-conditions to express fine-grained effects of potential library component candidates to drive a bi-directional synthesis search strategy. The procedure alternates between a forward search process that seeks to build larger terms given an existing context but which is otherwise unaware of the actual goal, alongside a backward search mechanism that seeks terms consistent with the desired goal but which is otherwise unaware of the context from which these terms must be synthesized. To further improve efficiency and scalability, we integrate a conflict-driven learning procedure into the synthesis algorithm that provides a semantic characterization of previously encountered unsuccessful search paths that is used to prune the space of possible candidates as synthesis proceeds. We have implemented our ideas in a tool called Cobalt and demonstrate its effectiveness on a number of challenging synthesis problems defined over OCaml libraries equipped with effectful specifications.

Deep learning networks have shown promising results in fast magnetic resonance imaging (MRI) reconstruction. In our work, we develop deep networks to further improve the quantitative and the perceptual quality of reconstruction. To begin with, we propose reconsynergynet (RSN), a network that combines the complementary benefits of independently operating on both the image and the Fourier domain. For a single-coil acquisition, we introduce deep cascade RSN (DC-RSN), a cascade of RSN blocks interleaved with data fidelity (DF) units. Secondly, we improve the structure recovery of DC-RSN for T2 weighted Imaging (T2WI) through assistance of T1 weighted imaging (T1WI), a sequence with short acquisition time. T1 assistance is provided to DC-RSN through a gradient of log feature (GOLF) fusion. Furthermore, we propose perceptual refinement network (PRN) to refine the reconstructions for better visual information fidelity (VIF), a metric highly correlated to radiologists opinion on the image quality. Lastly, for multi-coil acquisition, we propose variable splitting RSN (VS-RSN), a deep cascade of blocks, each block containing RSN, multi-coil DF unit, and a weighted average module. We extensively validate our models DC-RSN and VS-RSN for single-coil and multi-coil acquisitions and report the state-of-the-art performance. We obtain a SSIM of 0.768, 0.923, 0.878 for knee single-coil-4x, multi-coil-4x, and multi-coil-8x in fastMRI. We also conduct experiments to demonstrate the efficacy of GOLF based T1 assistance and PRN., Comment: Accepted in CMIG 2021

While notable progress has been made in specifying and learning objectives for general cyber-physical systems, applying these methods to distributed multi-agent systems still pose significant challenges. Among these are the need to (a) craft specification primitives that allow expression and interplay of both local and global objectives, (b) tame explosion in the state and action spaces to enable effective learning, and (c) minimize coordination frequency and the set of engaged participants for global objectives. To address these challenges, we propose a novel specification framework that allows natural composition of local and global objectives used to guide training of a multi-agent system. Our technique enables learning expressive policies that allow agents to operate in a coordination-free manner for local objectives, while using a decentralized communication protocol for enforcing global ones. Experimental results support our claim that sophisticated multi-agent distributed planning problems can be effectively realized using specification-guided learning., Comment: ECML-PKDD 2022

Neural network policies trained using Deep Reinforcement Learning (DRL) are well-known to be susceptible to adversarial attacks. In this paper, we consider attacks manifesting as perturbations in the observation space managed by the external environment. These attacks have been shown to downgrade policy performance significantly. We focus our attention on well-trained deterministic and stochastic neural network policies in the context of continuous control benchmarks subject to four well-studied observation space adversarial attacks. To defend against these attacks, we propose a novel defense strategy using a detect-and-denoise schema. Unlike previous adversarial training approaches that sample data in adversarial scenarios, our solution does not require sampling data in an environment under attack, thereby greatly reducing risk during training. Detailed experimental results show that our technique is comparable with state-of-the-art adversarial training approaches.

We study the fundamental limits of classical communication using quantum states that decohere as they traverse through a network of queues. We consider a network of Markovian queues, known as a Jackson network, with a single source or multiple sources and a single destination. Qubits are communicated through this network with inevitable buffering at intermediate nodes. We model each node as a `queue-channel,' wherein as the qubits wait in buffer, they continue to interact with the environment and suffer a waiting time-dependent noise. Focusing on erasures, we first obtain explicit classical capacity expressions for simple topologies such as tandem queue-channel and parallel queue-channel. Using these as building blocks, we characterize the classical capacity of a general quantum Jackson network with waiting time-dependent erasures. Throughout, we study two types of quantum networks, namely, (i) Repeater-assisted and (ii) Repeater-less. We also obtain optimal pumping rates and routing probabilities to maximize capacity in simple topologies. More broadly, our work quantifies the impact of delay-induced decoherence on the fundamental limits of classical communication over quantum networks., Comment: 8 pages

We report on the discovery and localization of fast radio bursts (FRBs) from the MeerTRAP project, a commensal fast radio transient-detection programme at MeerKAT in South Africa. Our hybrid approach combines a coherent search with an average field-of-view of 0.4 $\rm deg^{2}$ with an incoherent search utilizing a field-of-view of $\sim$1.27 $\rm deg^{2}$ (both at 1284~MHz). Here, we present results on the first three FRBs: FRB 20200413A (DM=1990.05 pc cm$^{-3}$), FRB 20200915A (DM=740.65 pc cm$^{-3}$), and FRB 20201123A (DM=433.55 pc cm$^{-3}$). FRB 20200413A was discovered only in the incoherent beam. FRB 20200915A (also discovered only in the incoherent beam) shows speckled emission in the dynamic spectrum which cannot be explained by interstellar scintillation in our Galaxy or plasma lensing, and might be intrinsic to the source. FRB 20201123A shows a faint post-cursor burst about 200 ms after the main burst and warrants further follow-up to confirm whether it is a repeating FRB. FRB 20201123A also exhibits significant temporal broadening consistent with scattering by a turbulent medium. The broadening exceeds that predicted for medium along the sightline through our Galaxy. We associate this scattering with the turbulent medium in the environment of the FRB in the host galaxy. Within the approximately $1'$ localization region of FRB 20201123A, we identify one luminous galaxy ($r \approx 15.67$; J173438.35$-$504550.4) that dominates the posterior probability for a host association. The galaxy's measured properties are consistent with other FRB hosts with secure associations., Comment: 14 pages, 9 figures, 3 tables, accepted for publication in MNRAS

We derive a new eight dimensional matrix representation of the Maxwell equations for a linear homogeneous medium and extend it to the case of a linear inhomogneous medium. This derivation starts ab initio with the Maxwell equations and uses arguments based on the algebra of the Pauli matrices. This process leads automatically to the matrix representation based on the Riemann-Silberstein-Weber (RSW) vector. The new representation for the homogeneous medium is a direct sum of four Pauli matrix blocks. This aspect of the new representation should make it suitable for studying the propagation of electromagnetic waves in a linear inhomogeneous medium adopting the techniques of quantum mechanics treating the inhomogeneity as a perturbation. The new representation is used to rederive the Mukunda-Simon-Sudarshan matrix substitution rule for transition from the Helmholtz scalar wave optics to the Maxwell vector wave optics., Comment: 50 pages. The previous version has been substantially revised. Journal Reference: Results in Optics (in press)

In several applications such as clinical trials and financial portfolio optimization, the expected value (or the average reward) does not satisfactorily capture the merits of a drug or a portfolio. In such applications, risk plays a crucial role, and a risk-aware performance measure is preferable, so as to capture losses in the case of adverse events. This survey aims to consolidate and summarise the existing research on risk measures, specifically in the context of multi-armed bandits. We review various risk measures of interest, and comment on their properties. Next, we review existing concentration inequalities for various risk measures. Then, we proceed to defining risk-aware bandit problems, We consider algorithms for the regret minimization setting, where the exploration-exploitation trade-off manifests, as well as the best-arm identification setting, which is a pure exploration problem -- both in the context of risk-sensitive measures. We conclude by commenting on persisting challenges and fertile areas for future research., Comment: 11 pages; Unabridged version of a a survey paper of the same title accepted to IJCAI-ECAI, 2022

Model-free Deep Reinforcement Learning (DRL) controllers have demonstrated promising results on various challenging non-linear control tasks. While a model-free DRL algorithm can solve unknown dynamics and high-dimensional problems, it lacks safety assurance. Although safety constraints can be encoded as part of a reward function, there still exists a large gap between an RL controller trained with this modified reward and a safe controller. In contrast, instead of implicitly encoding safety constraints with rewards, we explicitly co-learn a Twin Neural Lyapunov Function (TNLF) with the control policy in the DRL training loop and use the learned TNLF to build a runtime monitor. Combined with the path generated from a planner, the monitor chooses appropriate waypoints that guide the learned controller to provide collision-free control trajectories. Our approach inherits the scalability advantages from DRL while enhancing safety guarantees. Our experimental evaluation demonstrates the effectiveness of our approach compared to DRL with augmented rewards and constrained DRL methods over a range of high-dimensional safety-sensitive navigation tasks., Comment: 8 pages, 6 figures

We report the discovery of PSR J1555-2908, a 1.79 ms radio and gamma-ray pulsar in a 5.6 hr binary system with a minimum companion mass of 0.052 $M_\odot$. This fast and energetic ($\dot E = 3 \times 10^{35}$ erg/s) millisecond pulsar was first detected as a gamma-ray point source in Fermi LAT sky survey observations. Guided by a steep spectrum radio point source in the Fermi error region, we performed a search at 820 MHz with the Green Bank Telescope that first discovered the pulsations. The initial radio pulse timing observations provided enough information to seed a search for gamma-ray pulsations in the LAT data, from which we derive a timing solution valid for the full Fermi mission. In addition to the radio and gamma-ray pulsation discovery and timing, we searched for X-ray pulsations using NICER but no significant pulsations were detected. We also obtained time-series r-band photometry that indicates strong heating of the companion star by the pulsar wind. Material blown off the heated companion eclipses the 820 MHz radio pulse during inferior conjunction of the companion for ~10% of the orbit, which is twice the angle subtended by its Roche lobe in an edge-on system., Comment: 15 pages, 6 figures, accepted by ApJ

In this era of fast and large-scale opinion formation, a mathematical understanding of opinion evolution, a.k.a. opinion dynamics, is especially important. Linear graph-based dynamics and bounded confidence dynamics are the two most popular models for opinion dynamics in social networks. Recently, stochastic bounded confidence opinion dynamics were proposed as a general framework that incorporates both these dynamics as special cases and also captures the inherent stochasticity and noise (errors) in real-life social exchanges. Although these dynamics are quite general and realistic, their analysis is particularly challenging compared to other opinion dynamics models. This is because these dynamics are nonlinear and stochastic, and belong to the class of Markov processes that have asymptotically zero drift and unbounded jumps. The asymptotic behavior of these dynamics was characterized in prior works. However, they do not shed light on their finite-time behavior, which is often of interest in practice. We take a stride in this direction by analyzing the finite time behavior of a two-agent system, which is fundamental to the understanding of multi-agent dynamics. In particular, we show that the opinion difference between the two agents is well concentrated around zero under the conditions that lead to asymptotic stability of the dynamics., Comment: Accepted in COMSNETS 2022

In this work we performed a spectral energy distribution (SED) analysis in the optical/infrared band of the host galaxy of a proto-brightest cluster galaxy (BCG, NVSS J103023+052426) in a proto-cluster at z = 1.7. We found that it features a vigorous star formation rate (SFR) of ${\sim}$570 $\mathrm{M_{\odot}}$/yr and a stellar mass of $M_{\ast} \sim 3.7 \times 10^{11}$ $\mathrm{M_{\odot}}$; the high corresponding specific SFR = $1.5 \pm 0.5$ $\mathrm{Gyr^{-1}}$ classifies this object as a starburst galaxy that will deplete its molecular gas reservoir in $\sim$ $3.5 \times 10^8$ yr. Thus, this system represents a rare example of a proto-BCG caught during the short phase of its major stellar mass assembly. Moreover, we investigated the nature of the host galaxy emission at 3.3 mm. We found that it originates from the cold dust in the interstellar medium, even though a minor non-thermal AGN contribution cannot be completely ruled out. Finally, we studied the polarized emission of the lobes at 1.4 GHz. We unveiled a patchy structure where the polarization fraction increases in the regions in which the total intensity shows a bending morphology; in addition, the magnetic field orientation follows the direction of the bendings. We interpret these features as possible indications of an interaction with the intracluster medium. This strengthens the hypothesis of positive AGN feedback, as inferred in previous studies of this object on the basis of X-ray/mm/radio analysis. In this scenario, the proto-BCG heats the surrounding medium and possibly enhances the SFR in nearby galaxies., Comment: 13 pages, 3 figures, 1 table. Accepted for publication in Galaxies

We demonstrate an optical receiver that achieves the quantum Chernoff bound for discriminating coherent states from thermal states in the multi-copy scenario. In contrast, we find that repeated use of the receiver approaching the Helstrom bound for single-copy measurement is sub-optimal in this multi-copy case. Furthermore, for a large class of multi-copy discrimination tasks between a pure and a mixed state, we prove that any Helstrom-bound achieving single-copy receiver is suboptimal by a factor of at least two in error-probability exponent compared to the multi-copy quantum Chernoff bound. This behavior has a classical analog in the performance gap between soft-decision and hard-decision receivers for detecting a multi-copy signal embedded in white Gaussian noise., Comment: 11 pages, 4 figures

Utility-Based Shortfall Risk (UBSR) is a risk metric that is increasingly popular in financial applications, owing to certain desirable properties that it enjoys. We consider the problem of estimating UBSR in a recursive setting, where samples from the underlying loss distribution are available one-at-a-time. We cast the UBSR estimation problem as a root finding problem, and propose stochastic approximation-based estimations schemes. We derive non-asymptotic bounds on the estimation error in the number of samples. We also consider the problem of UBSR optimization within a parameterized class of random variables. We propose a stochastic gradient descent based algorithm for UBSR optimization, and derive non-asymptotic bounds on its convergence.

In this paper, we address two key challenges in deep reinforcement learning setting, sample inefficiency and slow learning, with a dual NN-driven learning approach. In the proposed approach, we use two deep NNs with independent initialization to robustly approximate the action-value function in the presence of image inputs. In particular, we develop a temporal difference (TD) error-driven learning approach, where we introduce a set of linear transformations of the TD error to directly update the parameters of each layer in the deep NN. We demonstrate theoretically that the cost minimized by the error-driven learning (EDL) regime is an approximation of the empirical cost and the approximation error reduces as learning progresses, irrespective of the size of the network. Using simulation analysis, we show that the proposed methods enables faster learning and convergence and requires reduced buffer size (thereby increasing the sample efficiency).

Learning to control complex systems using non-traditional feedback, e.g., in the form of snapshot images, is an important task encountered in diverse domains such as robotics, neuroscience, and biology (cellular systems). In this paper, we present a two neural-network (NN)-based feedback control framework to design control policies for systems that generate feedback in the form of images. In particular, we develop a deep $Q$-network (DQN)-driven learning control strategy to synthesize a sequence of control inputs from snapshot images that encode the information pertaining to the current state and control action of the system. Further, to train the networks we employ a direct error-driven learning (EDL) approach that utilizes a set of linear transformations of the NN training error to update the NN weights in each layer. We verify the efficacy of the proposed control strategy using numerical examples.

Semantic mapping is the task of providing a robot with a map of its environment beyond the open, navigable space of traditional Simultaneous Localization and Mapping (SLAM) algorithms by attaching semantics to locations. The system presented in this work reads door placards to annotate the locations of offices. Whereas prior work on this system developed hand-crafted detectors, this system leverages YOLOv2 for detection and a segmentation network for segmentation. Placards are localized by computing their pose from a homography computed from a segmented quadrilateral outline. This work also introduces an Interruptable Frontier Exploration algorithm, enabling the robot to explore its environment to construct its SLAM map while pausing to inspect placards observed during this process. This allows the robot to autonomously discover room placards without human intervention while speeding up significantly over previous autonomous exploration methods.

Quantum queue-channels arise naturally in the context of buffering in quantum networks, wherein the noise suffered by the quantum states depends on the time spent waiting in the buffer. It has been shown that the upper-bound on the classical capacity of an additive queue-channel has a simple expression, and is achievable for the erasure and depolarizing channels [IEEE JSAIT, 1(2):432-444]. In this paper, we characterise the classical capacity for the class of unital qubit queue-channels, and show that a simple product (non-entangled) decoding strategy is capacity-achieving. As an intermediate result, we derive an explicit capacity achieving product decoding strategy for any i.i.d. unital qubit channel, which could be of independent interest. As an important special case, we also derive the capacity and optimal decoding strategies for a symmetric generalized amplitude damping (GAD) queue-channel. Our results provide useful insights towards designing practical quantum communication networks, and highlight the need to explicitly model the impact of buffering., Comment: 25 pages with 3 Figures. Merges and supersedes our prior submissions arXiv:2107.13486 and arXiv:2110.02800

Fine-tuning pretrained models for automatically summarizing doctor-patient conversation transcripts presents many challenges: limited training data, significant domain shift, long and noisy transcripts, and high target summary variability. In this paper, we explore the feasibility of using pretrained transformer models for automatically summarizing doctor-patient conversations directly from transcripts. We show that fluent and adequate summaries can be generated with limited training data by fine-tuning BART on a specially constructed dataset. The resulting models greatly surpass the performance of an average human annotator and the quality of previous published work for the task. We evaluate multiple methods for handling long conversations, comparing them to the obvious baseline of truncating the conversation to fit the pretrained model length limit. We introduce a multistage approach that tackles the task by learning two fine-tuned models: one for summarizing conversation chunks into partial summaries, followed by one for rewriting the collection of partial summaries into a complete summary. Using a carefully chosen fine-tuning dataset, this method is shown to be effective at handling longer conversations, improving the quality of generated summaries. We conduct both an automatic evaluation (through ROUGE and two concept-based metrics focusing on medical findings) and a human evaluation (through qualitative examples from literature, assessing hallucination, generalization, fluency, and general quality of the generated summaries)., Comment: Accepted in Findings of the EMNLP 2021. Code is available at https://github.com/negrinho/medical_conversation_summarization

Programmers often leverage data structure libraries that provide useful and reusable abstractions. Modular verification of programs that make use of these libraries naturally rely on specifications that capture important properties about how the library expects these data structures to be accessed and manipulated. However, these specifications are often missing or incomplete, making it hard for clients to be confident they are using the library safely. When library source code is also unavailable, as is often the case, the challenge to infer meaningful specifications is further exacerbated. In this paper, we present a novel data-driven abductive inference mechanism that infers specifications for library methods sufficient to enable verification of the library's clients. Our technique combines a data-driven learning-based framework to postulate candidate specifications, along with SMT-provided counterexamples to refine these candidates, taking special care to prevent generating specifications that overfit to sampled tests. The resulting specifications form a minimal set of requirements on the behavior of library implementations that ensures safety of a particular client program. Our solution thus provides a new multi-abduction procedure for precise specification inference of data structure libraries guided by client-side verification tasks. Experimental results on a wide range of realistic OCaml data structure programs demonstrate the effectiveness of the approach.