BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Sukhdeep Singh

Sukhdeep Singh contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
astro-ph.COastro-ph.GAComputer VisionMachine LearningArtificial Intelligenceeess.SPNetworking and Internet Architectureastro-ph.IMComputation and Languagehep-exMultiagent Systems

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
15works
0followers
11topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

15 published item(s)

preprint2026arXiv

Towards Conversational Medical AI with Eyes, Ears and a Voice

The practice of medicine relies not only upon skillful dialogue but also on the nuanced exchange and interpretation of rich auditory and visual cues between doctors and patients. Building on the low-latency voice and video processing capabilities of Gemini, we introduce AI co-clinician, a first-of-its-kind conversational AI system utilizing continuous streams of audio-visual data from live patient conversations to inform real-time clinical decisions. Its dual-agent architecture balances deep clinical reasoning with the low latency required for natural dialogue. To assess this system, we implemented a video-based interface emulating telemedicine consultations. We crafted 20 standardized outpatient scenarios requiring proactive real-time auditory and visual reasoning and designed "TelePACES" evaluation criteria alongside case-specific rubrics. In a randomized, interface-blinded, crossover simulation study (n = 120 encounters) with 10 internal medicine residents as patient actors, we compared AI co-clinician with primary care physicians (PCPs), GPT-Realtime, and a baseline agent. AI co-clinician approached PCPs in key TelePACES dimensions, including management plans and differential diagnosis, while significantly outperforming GPT-Realtime across all general criteria. While our agent demonstrated parity with PCPs in case-specific triage measures, physicians maintained superior overall performance in case-specific assessments. Although AI co-clinician marks a significant advance in real-time telemedical AI, gaps remain in physical examination and disease-specific reasoning. Our work shows that text-only approaches fail to capture the true challenges of medical consultation and suggests that high-stakes real-time diagnostic AI is most safely advanced in collaborative, triadic models where AI can be a supportive co-clinician for doctors and patients.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Lexicon and Depth-wise Separable Convolution Based Handwritten Text Recognition System

Cursive handwritten text recognition is a challenging research problem in the domain of pattern recognition. The current state-of-the-art approaches include models based on convolutional recurrent neural networks and multi-dimensional long short-term memory recurrent neural networks techniques. These methods are highly computationally extensive as well model is complex at design level. In recent studies, combination of convolutional neural network and gated convolutional neural networks based models demonstrated less number of parameters in comparison to convolutional recurrent neural networks based models. In the direction to reduced the total number of parameters to be trained, in this work, we have used depthwise convolution in place of standard convolutions with a combination of gated-convolutional neural network and bidirectional gated recurrent unit to reduce the total number of parameters to be trained. Additionally, we have also included a lexicon based word beam search decoder at testing step. It also helps in improving the the overall accuracy of the model. We have obtained 3.84% character error rate and 9.40% word error rate on IAM dataset; 4.88% character error rate and 14.56% word error rate in George Washington dataset, respectively.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Galaxies and Halos on Graph Neural Networks: Deep Generative Modeling Scalar and Vector Quantities for Intrinsic Alignment

In order to prepare for the upcoming wide-field cosmological surveys, large simulations of the Universe with realistic galaxy populations are required. In particular, the tendency of galaxies to naturally align towards overdensities, an effect called intrinsic alignments (IA), can be a major source of systematics in the weak lensing analysis. As the details of galaxy formation and evolution relevant to IA cannot be simulated in practice on such volumes, we propose as an alternative a Deep Generative Model. This model is trained on the IllustrisTNG-100 simulation and is capable of sampling the orientations of a population of galaxies so as to recover the correct alignments. In our approach, we model the cosmic web as a set of graphs, where the graphs are constructed for each halo, and galaxy orientations as a signal on those graphs. The generative model is implemented on a Generative Adversarial Network architecture and uses specifically designed Graph-Convolutional Networks sensitive to the relative 3D positions of the vertices. Given (sub)halo masses and tidal fields, the model is able to learn and predict scalar features such as galaxy and dark matter subhalo shapes; and more importantly, vector features such as the 3D orientation of the major axis of the ellipsoid and the complex 2D ellipticities. For correlations of 3D orientations the model is in good quantitative agreement with the measured values from the simulation, except for at very small and transition scales. For correlations of 2D ellipticities, the model is in good quantitative agreement with the measured values from the simulation on all scales. Additionally, the model is able to capture the dependence of IA on mass, morphological type and central/satellite type.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Intelligent Ranking for Dynamic Restoration in Next Generation Wireless Networks

Emerging 5G and next generation 6G wireless are likely to involve myriads of connectivity, consisting of a huge number of relatively smaller cells providing ultra-dense coverage. Guaranteeing seamless connectivity and service level agreements in such a dense wireless system demands efficient network management and fast service recovery. However, restoration of a wireless network, in terms of maximizing service recovery, typically requires evaluating the service impact of every network element. Unfortunately, unavailability of real-time KPI information, during an outage, enforces most of the existing approaches to rely significantly on context-based manual evaluation. As a consequence, configuring a real-time recovery of the network nodes is almost impossible, thereby resulting in a prolonged outage duration. In this article, we explore deep learning to introduce an intelligent, proactive network recovery management scheme in anticipation of an eminent network outage. Our proposed method introduces a novel utilization-based ranking scheme of different wireless nodes to minimize the service downtime and enable a fast recovery. Efficient prediction of network KPI (Key Performance Index), based on actual wireless data demonstrates up to ~54% improvement in service outage.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Intrinsic alignments of bulges and discs

Galaxies exhibit coherent alignments with local structure in the Universe. This effect, called Intrinsic Alignments (IA), is an important contributor to the systematic uncertainties for wide-field weak lensing surveys. On cosmological distance scales, intrinsic shape alignments have been observed in red galaxies, which are usually bulge-dominated; while blue galaxies, which are mostly disc-dominated, exhibit shape alignments consistent with a null detection. However, disc-dominated galaxies typically consist of two prominent structures: disc and bulge. Since the bulge component has similar properties as elliptical galaxies and is thought to have formed in a similar fashion, naturally one could ask whether the bulge components exhibit similar alignments as ellipticals? In this paper, we investigate how different components of galaxies exhibit IA in the TNG100-1 cosmological hydrodynamical simulation, as well as the dependence of IA on the fraction of stars in rotation-dominated structures at $z=0$. The measurements were controlled for mass differences between the samples. We find that the bulges exhibit significantly higher IA signals, with a nonlinear alignment model amplitude of $A_I = 2.98^{+0.36}_{-0.37}$ compared to the amplitude for the galaxies as a whole (both components), $A_I = 1.13^{+0.37}_{-0.35}$. The results for bulges are statistically consistent with those for elliptical galaxies, which have $A_I = 3.47^{+0.57}_{-0.57}$. These results highlight the importance of studying galaxy dynamics in order to understand galaxy alignments and their cosmological implications.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Learning Robust Real-Time Cultural Transmission without Human Data

Cultural transmission is the domain-general social skill that allows agents to acquire and use information from each other in real-time with high fidelity and recall. In humans, it is the inheritance process that powers cumulative cultural evolution, expanding our skills, tools and knowledge across generations. We provide a method for generating zero-shot, high recall cultural transmission in artificially intelligent agents. Our agents succeed at real-time cultural transmission from humans in novel contexts without using any pre-collected human data. We identify a surprisingly simple set of ingredients sufficient for generating cultural transmission and develop an evaluation methodology for rigorously assessing it. This paves the way for cultural evolution as an algorithm for developing artificial general intelligence.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Lexicon and Attention based Handwritten Text Recognition System

The handwritten text recognition problem is widely studied by the researchers of computer vision community due to its scope of improvement and applicability to daily lives, It is a sub-domain of pattern recognition. Due to advancement of computational power of computers since last few decades neural networks based systems heavily contributed towards providing the state-of-the-art handwritten text recognizers. In the same direction, we have taken two state-of-the art neural networks systems and merged the attention mechanism with it. The attention technique has been widely used in the domain of neural machine translations and automatic speech recognition and now is being implemented in text recognition domain. In this study, we are able to achieve 4.15% character error rate and 9.72% word error rate on IAM dataset, 7.07% character error rate and 16.14% word error rate on GW dataset after merging the attention and word beam search decoder with existing Flor et al. architecture. To analyse further, we have also used system similar to Shi et al. neural network system with greedy decoder and observed 23.27% improvement in character error rate from the base model.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Predictive Closed-Loop Service Automation in O-RAN based Network Slicing

Network slicing provides introduces customized and agile network deployment for managing different service types for various verticals under the same infrastructure. To cater to the dynamic service requirements of these verticals and meet the required quality-of-service (QoS) mentioned in the service-level agreement (SLA), network slices need to be isolated through dedicated elements and resources. Additionally, allocated resources to these slices need to be continuously monitored and intelligently managed. This enables immediate detection and correction of any SLA violation to support automated service assurance in a closed-loop fashion. By reducing human intervention, intelligent and closed-loop resource management reduces the cost of offering flexible services. Resource management in a network shared among verticals (potentially administered by different providers), would be further facilitated through open and standardized interfaces. Open radio access network (O-RAN) is perhaps the most promising RAN architecture that inherits all the aforementioned features, namely intelligence, open and standard interfaces, and closed control loop. Inspired by this, in this article we provide a closed-loop and intelligent resource provisioning scheme for O-RAN slicing to prevent SLA violations. In order to maintain realism, a real-world dataset of a large operator is used to train a learning solution for optimizing resource utilization in the proposed closed-loop service automation process. Moreover, the deployment architecture and the corresponding flow that are cognizant of the O-RAN requirements are also discussed.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Snowmass2021 Cosmic Frontier White Paper: Enabling Flagship Dark Energy Experiments to Reach their Full Potential

A new generation of powerful dark energy experiments will open new vistas for cosmology in the next decade. However, these projects cannot reach their utmost potential without data from other telescopes. This white paper focuses in particular on the compelling benefits of ground-based spectroscopic and photometric observations to complement the Vera C. Rubin Observatory, as well as smaller programs in aid of a DESI-2 experiment and CMB-S4. These additional data sets will both improve dark energy constraints from these flagship projects beyond what would possible on their own and open completely new windows into fundamental physics. For example, additional photometry and single-object spectroscopy will provide necessary follow-up information for supernova and strong lensing cosmology, while highly-multiplexed spectroscopy both from smaller facilities over wide fields and from larger facilities over narrower regions of sky will yield more accurate photometric redshift estimates for weak lensing and galaxy clustering measurements from the Rubin Observatory, provide critical spectroscopic host galaxy redshifts for supernova Hubble diagrams, provide improved understanding of limiting astrophysical systematic effects, and enable new measurements that probe the nature of gravity. A common thread is that access to complementary data from a range of telescopes/instruments would have a substantial impact on the rate of advance of dark energy science in the coming years.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

On the halo-mass and radial scale dependence of the lensing is low effect

The canonical $Λ$CDM cosmological model makes precise predictions for the clustering and lensing properties of galaxies. It has been shown that the lensing amplitude of galaxies in the Baryon Oscillation Spectroscopic Survey (BOSS) is lower than expected given their clustering properties. We present new measurements and modelling of galaxies in the BOSS LOWZ sample. We focus on the radial and stellar mass dependence of the lensing amplitude mis-match. We find an amplitude mis-match of around $35\%$ when assuming $Λ$CDM with Planck Cosmological Microwave Background (CMB) constraints. This offset is independent of halo mass and radial scale in the range $M_{\rm halo}\sim 10^{13.3} - 10^{13.9} h^{-1} M_\odot$ and $r=0.1 - 60 \, h^{-1} \mathrm{Mpc}$ ($k \approx 0.05 - 20 \, h \, {\rm Mpc}^{-1}$). The observation that the offset is both mass and scale independent places important constraints on the degree to which astrophysical processes (baryonic effects, assembly bias) can fully explain the effect. This scale independence also suggests that the "lensing is low" effect on small and large radial scales probably have the same physical origin. Resolutions based on new physics require a nearly uniform suppression, relative to $Λ$CDM predictions, of the amplitude of matter fluctuations on these scales. The possible causes of this are tightly constrained by measurements of the CMB and of the low-redshift expansion history.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Cosmological constraints from galaxy-lensing cross correlations using BOSS galaxies with SDSS and CMB lensing

We present cosmological parameter constraints based on a joint modeling of galaxy-lensing cross correlations and galaxy clustering measurements in the SDSS, marginalizing over small-scale modeling uncertainties using mock galaxy catalogs, without explicit modeling of galaxy bias. We show that our modeling method is robust to the impact of different choices for how galaxies occupy dark matter halos and to the impact of baryonic physics (at the $\sim2\%$ level in cosmological parameters) and test for the impact of covariance on the likelihood analysis and of the survey window function on the theory computations. Applying our results to the measurements using galaxy samples from BOSS and lensing measurements using shear from SDSS galaxies and CMB lensing from Planck, with conservative scale cuts, we obtain $S_8\equiv\left(\frac{σ_8}{0.8228}\right)^{0.8}\left(\frac{Ω_m}{0.307}\right)^{0.6}=0.85\pm0.05$ (stat.) using LOWZ $\times$ SDSS galaxy lensing, and $S_8=0.91\pm0.1$ (stat.) using combination of LOWZ and CMASS $\times$ Planck CMB lensing. We estimate the systematic uncertainty in the galaxy-galaxy lensing measurements to be $\sim6\%$ (dominated by photometric redshift uncertainties) and in the galaxy-CMB lensing measurements to be $\sim3\%$, from small scale modeling uncertainties including baryonic physics.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Intelligent O-RAN for Beyond 5G and 6G Wireless Networks

Building on the principles of openness and intelligence, there has been a concerted global effort from the operators towards enhancing the radio access network (RAN) architecture. The objective is to build an operator-defined RAN architecture (and associated interfaces) on open hardware that provides intelligent radio control for beyond fifth generation (5G) as well as future sixth generation (6G) wireless networks. Specifically, the open-radio access network (O-RAN) alliance has been formed by merging xRAN forum and C-RAN alliance to formally define the requirements that would help achieve this objective. Owing to the importance of O-RAN in the current wireless landscape, this article provides an introduction to the concepts, principles, and requirements of the Open RAN as specified by the O-RAN alliance. In order to illustrate the role of intelligence in O-RAN, we propose an intelligent radio resource management scheme to handle traffic congestion and demonstrate its efficacy on a real-world dataset obtained from a large operator. A high-level architecture of this deployment scenario that is compliant with the O-RAN requirements is also discussed. The article concludes with key technical challenges and open problems for future research and development.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

Core Cosmology Library: Precision Cosmological Predictions for LSST

The Core Cosmology Library (CCL) provides routines to compute basic cosmological observables to a high degree of accuracy, which have been verified with an extensive suite of validation tests. Predictions are provided for many cosmological quantities, including distances, angular power spectra, correlation functions, halo bias and the halo mass function through state-of-the-art modeling prescriptions available in the literature. Fiducial specifications for the expected galaxy distributions for the Large Synoptic Survey Telescope (LSST) are also included, together with the capability of computing redshift distributions for a user-defined photometric redshift model. A rigorous validation procedure, based on comparisons between CCL and independent software packages, allows us to establish a well-defined numerical accuracy for each predicted quantity. As a result, predictions for correlation functions of galaxy clustering, galaxy-galaxy lensing and cosmic shear are demonstrated to be within a fraction of the expected statistical uncertainty of the observables for the models and in the range of scales of interest to LSST. CCL is an open source software package written in C, with a python interface and publicly available at https://github.com/LSSTDESC/CCL.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

Cosmology with galaxy-galaxy lensing on non-perturbative scales: Emulation method and application to BOSS LOWZ

We describe our nonlinear emulation (i.e., interpolation) framework that combines the halo occupation distribution (HOD) galaxy bias model with $N$-body simulations of nonlinear structure formation, designed to accurately predict the projected clustering and galaxy-galaxy lensing signals from luminous red galaxies (LRGs) in the redshift range $0.16 < z < 0.36$ on comoving scales $0.6 < r_p < 30$ \hMpc. The interpolation accuracy is $\lesssim 1-2$ per cent across the entire physically plausible range of parameters for all scales considered. We correctly recover the true value of the cosmological parameter $S_8 = ({σ_8}/{0.8228}) ({Ω_{\text{m}}}/{0.3107})^{0.6}$ from mock measurements produced via subhalo abundance matching (SHAM)-based lightcones designed to approximately match the properties of the SDSS LOWZ galaxy sample. Applying our model to Baryon Oscillation Spectroscopic Survey (BOSS) Data Release 14 (DR14) LOWZ galaxy clustering and galaxy-shear cross-correlation measurements made with Sloan Digital Sky Survey (SDSS) Data Release 8 (DR8) imaging, we perform a prototype cosmological analysis marginalizing over $w$CDM cosmological parameters and galaxy HOD parameters. We obtain a 4.4 per cent measurement of $S_8 = 0.847 \pm 0.037$, in $3.5σ$ tension with the Planck cosmological results of $1.00 \pm 0.02$. We discuss the possibility of underestimated systematic uncertainties or astrophysical effects that could explain this discrepancy.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

High mass and halo resolution from fast low resolution simulations

Generating mocks for future sky surveys requires large volumes and high resolutions, which is computationally expensive even for fast simulations. In this work we try to develop numerical schemes to calibrate various halo and matter statistics in fast low resolution simulations compared to high resolution N-body and hydrodynamic simulations. For the halos, we improve the initial condition accuracy and develop a halo finder &#34;relaxed-FOF&#34;, where we allow different linking length for different halo mass and velocity dispersions. We show that our relaxed-FoF halo finder improves the common statistics, such as halo bias, halo mass function, halo auto power spectrum in real space and in redshift space, cross correlation coefficient with the reference halo catalog, and halo-matter cross power spectrum. We also incorporate the potential gradient descent (PGD) method into fast simulations to improve the matter distribution at nonlinear scale. By building a lightcone output, we show that the PGD method significantly improves the weak lensing convergence tomographic power spectrum. With these improvements FastPM is comparable to the high resolution full N-body simulation of the same mass resolution, with two orders of magnitude fewer time steps. These techniques can be used to improve the halo and matter statistics of FastPM simulations for mock catalogs of future surveys such as DESI and LSST.

0 citations0 reviewsNo code linkedOpen access