Transfer Learning for Human & AI

This paper gives a systematic research review at the boundary of theinformation systems (IS) and marketing disciplines . The citation asymmetries between IS and marketing are noted and an overall conceptual model is created . Forwardlooking suggestions are made on how academic researchers can better interfacewith industry and how academic research at the boundaries of IS can be further developed .…

Resilience is a concept of rising interest in computer science and software engineering . For systems in which correctness w.r.t. a safety condition is impossible, fast recovery is demanded . We investigate resilience problems ofgraph transformation systems . Our main contribution is the decidability of tworesilience problems for well-structured graph transformation systems (withstrong compatibility)…

This paper gives a systematic research review at the boundary of theinformation systems (IS) and marketing disciplines . The citation asymmetries between IS and marketing are noted and an overall conceptual model is created . Forwardlooking suggestions are made on how academic researchers can better interfacewith industry and how academic research at the boundaries of IS can be further developed .…

This paper gives a systematic research review at the boundary of theinformation systems (IS) and marketing disciplines . The citation asymmetries between IS and marketing are noted and an overall conceptual model is created . Forwardlooking suggestions are made on how academic researchers can better interfacewith industry and how academic research at the boundaries of IS can be further developed .…

A vast area of research in historical science concerns the documentation and study of artefacts and related evidence . Current practice mostly usesspreadsheets or simple relational databases to organise the information as rowswith multiple columns of related attributes . Synthesis documentationsystem is Web-based andcollaborative, and makes use of existing standards for informationdocumentation and publication (CIDOC-CRM, RDF), focusing on semanticinteroperability and the production of data of high value and long-termvalidity .…

Chemical reaction networks are formalized as chemical reaction networks (CRNs) Despite widespread use of CRNs in the natural sciences, the range of computational behavior exhibited by CRNs is not well understood . Understanding the algorithmic behaviors that are *in principle* realizable in a chemical system is necessary for a rigorous understanding of the designprinciples of biological regulatory networks .…

In model checking, when a given model fails to satisfy the desiredspecification, a typical model checker provides a counterexample that shows how the violation occurs . In general, there exist many diversecounterexamples that exhibit distinct violating behaviors . The goal of classification is to partition thespace of all countereXamples into a finite set of classes, each of which describes a distinct type of violating behavior for the givenspecification .…

Ten years ago, the Mathematics Subject Classification MSC 2010 was released . Now, the new MSC 2020 isout . This paper recaps the last ten years of working on machine-readable MSC data . We describe the processing required to convert the version of record into the Linked Open Data form we call MSC2020-SKOS .…

PHiLIP is an open-source externalreference device together with tools that validate the system software while it controls the hardware and interprets physical signals . The project’s initial deployment comprises 22 uniqueplatforms, each of which executes 98 peripheral tests every night .…

Quantum annealers manufactured by D-Wave Systems, Inc., are computational devices capable of finding high-quality solutions of NP-hard problems . In this contribution, we explore the potential and effectiveness of such quantumanneals for computing Boolean tensor networks . We show that tensor with up to millions of elements can bedecomposed efficiently using a DWave 2000Q adiabatic quantum annealer .…

Quantitative finance is the use of mathematical models to analyse financial markets and securities . We explore redesigning the engine to fully embrace the dataflow approach, resulting in an engine which is around eight times faster on anAlveo U280 FPGA than the original Xilinx library version .…

This article maps Library and Information Science (LIS) research in Indian universities . Web of Science and Scopus have very limited coverage of Indian LIS journals . Study traces growth of publications, prominent keywords, leading journals where articles are published .…

In this paper, we extend the notion of (word) derivatives and partialderivatives due to (respectively) Brzozowski and Antimirov to tree derivatives . We define a new family of extended regular tree expressions (using negationor intersection operators) We show how to compute a Brozozowski-likeinductive tree automaton .…

Current Python programming environment does not have any reliable and efficient multiple precision floating-point arithmetic . MPF library can be utilized for middlelength precision arithmetic under 200 bits, but they are not widely used on Python . In this paper, we describe our accelerated MPF direct method withAVX2 techniques and its application to mixed precision iterative refinement combined with mpmath, and demonstrate their efficiency on x86\_64 computational environments .…

Quantitative finance is the use of mathematical models to analyse financial markets and securities . We explore redesigning the engine to fully embrace the dataflow approach, resulting in an engine which is around eight times faster on anAlveo U280 FPGA than the original Xilinx library version .…

Current Python programming environment does not have any reliable and efficient multiple precision floating-point arithmetic . MPF library can be utilized for middlelength precision arithmetic under 200 bits, but they are not widely used on Python . In this paper, we describe our accelerated MPF direct method withAVX2 techniques and its application to mixed precision iterative refinement combined with mpmath, and demonstrate their efficiency on x86\_64 computational environments .…

Fifty years of evolution of the transportation field is revisited at a macroscale using scientometric analysis of all publications in all 39 journals indexed in the category of Transportation by the Web of Science . Influential and emerging authors of each division are identified .…

Nanoelectronic devices emulating neuro-synaptic functionalities through theirintrinsic physics at low operating energies is imperative toward therealization of brain-like neuromorphic computers . We leverage thenon-linear voltage dependent partial polarization switching of a ferroelectricfield effect transistor to mimic plasticity characteristics of biological synapses .…

Slovenia was the most productive country, followed byCroatia and Serbia . Nursing research in ex-Yugoslavian countries is growing both in scope and number of publications . A substantial part of the research is published in national journalsin national languages . The study also revealed substantial international cooperation among ex-yugoslavians and European Union.…

In this paper we explore the challenges of automating experiments in datascience . We propose an extensible experiment model as a foundation forintegration of different open source tools for running research experiments . Weimplement our approach in a prototype open source MLDev software package andevaluate it in a series of experiments yielding promising results .…

In this paper, we study the extension of 1-clock Alternating Timed Automata(1-ATA) with the ability to read in both forward and backward direction . We propose a “non-punctuality” like restriction,called non-adjacency, for 2-Way 1-ATA-rfl, and also for GQMSO, for which theemptiness (respectively, satisfiability) checking becomes decidable, in general .…

In this paper we explore the challenges of automating experiments in datascience . We propose an extensible experiment model as a foundation forintegration of different open source tools for running research experiments . Weimplement our approach in a prototype open source MLDev software package andevaluate it in a series of experiments yielding promising results .…

Reconfigurable Intelligent Surfaces (RISs) have attracted attention in the last year for their characteristics of nearly-passive, thin structures that candynamically change their reflection or refraction behaviour . In this paper, we introduce a general macroscopic model for therealistic evaluation of RIS scattering, based on its decomposition intomultiple scattering mechanisms and aimed at being embedded into ray models .…

We suggest an enhancement to structural coding through the use of (a)causally bound codes, (b) basic constructs of graph theory and (c) statistics . The numberof occurrences of the nodes and the links provide the input required to analyzerelative frequency of occurrence, as well as opening a scope for furtherstatistical analysis .…

Machine Learning (ML) has increasingly become a hottopic in the field of Computer Networks . During 2020, the International Telecommunication Union (ITU) hasorganized the “ITU AI/ML in 5G challenge”, an open global competition that has introduced to a broad audience some of the current main challenges in ML fornetworks .…

Machine Learning (ML) has increasingly become a hottopic in the field of Computer Networks . During 2020, the International Telecommunication Union (ITU) hasorganized the “ITU AI/ML in 5G challenge”, an open global competition that has introduced to a broad audience some of the current main challenges in ML fornetworks .…

Novel methods are presented for the fusion of GPU kernels in the artificialcompressibility method (ACM) The fused kernels are able to achieve 3-4 times speedup, which compares favourably with atheoretical maximum speedup of 4.3 times . In three dimensional test cases, thegenerated fused kernels were found to reduce total runtime by ${\sim}25\%$, and,when compared to the standard ACM formulation, simulations demonstrate that aspeedup of $2.3$ times can be achieved .…

Novel methods are presented for the fusion of GPU kernels in the artificialcompressibility method (ACM) The fused kernels are able to achieve 3-4 times speedup, which compares favourably with atheoretical maximum speedup of 4.3 times . In three dimensional test cases, thegenerated fused kernels were found to reduce total runtime by ${\sim}25\%$, and,when compared to the standard ACM formulation, simulations demonstrate that aspeedup of $2.3$ times can be achieved .…

Design of microfluidic biochips has led to newer challenges to the EDA community due to the availability of various flow-based architectures and the need for catering to diverse applications such as sample preparation, personalized medicine, point-of-care diagnostics, and drug design .…

Journal classification into subject categories is an important aspect inscholarly research evaluation as well as in bibliometric analysis . Web Of Science and Scopus indexing systems use a variety of (partially) overlapping and non-exhaustive subject categories which results in many journals being classified into more than a single subject category .…

Adiabatic quantum-flux-parametron (AQFP) logic is an energy-efficientuperconductor logic family . The latency of AQFP circuits is relatively long compared to that of other superconductor Logic families . And and XORgates with a latency of approximately 20 ps per gate are shown to operate at upto 5 and 4 GHz, respectively, in experiments .…

An uninterpreted program (UP) is a program whose semantics is defined over the theory of uninterpretated functions . This is a common abstraction used inequivalence checking, compiler optimization, and program verification . Recently, a class of UP programs, called coherent, has been proposed and shown to be decidable .…

LightOn’s Optical Processing Unit (OPU) is the first photonic AIaccelerator chip available on the market for at-scale Non von Neumanncomputations . It relies on a combination of free-spaceoptics with off-the-shelf components, together with a software API allowing aseamless integration within Python-based processing pipelines .…

Neuromorphic vision sensors (NVS) can enable energy savings due to theirevent-driven that exploits the temporal redundancy in video streams from astationary camera . Noise-driven events lead to the false triggering of the object recognition processor . Image denoise operations require memoryintensive processing leading to a bottleneck in energy and latency .…

This research note tries to uncover the 1991 phenomenon from theperspective of database limitation by probing the limitations of search inabstract/author keywords/keywords plus fields of Web of Science Core Collection . The lowavailability rates of abstract/author keyword searches in WoSCC found in this study can explain the “watershed” phenomenon of AIscholarship in 1991 to a large extent .…

SuperCaustics is a real-time,open-source simulation of transparent objects designed for deep learning applications . It uses hardware ray-tracing to support caustics, dispersion, and refraction . The neural network achieved performance comparable to the state-of-the-art on a real world dataset using only 10% of the training data and in a fraction of training time .…

Point clouds acquired from scanning devices are often perturbed by noise, which affects downstream tasks such as surface reconstruction and analysis . To denoise a noisy point cloud, we propose to increase the log-likelihood of each point from $p *n$ via gradient ascent .…

Recent state-of-the-art active learning methods have mostly leveragedGenerative Adversarial Networks (GAN) for sample acquisition . However, GAN isusually known to suffer from instability and sensitivity to hyper-parameters . In contrast to these methods, we propose in this paper a novel active learningframework that we call Maximum Classifier Discrepancy for Active Learning(MCDAL) which takes the prediction discrepancies between multiple classifiers .…

CompressiveSensing Imaging (CSI) is a challenging problem and has many practical applications . Deep neural networks have been applied to this problem with promising results . But existing neural network approaches require separate models for each imaging parameter like sampling ratios, leading totraining difficulties and overfitting to specific settings .…

A novel shape-perception network (HSPN) is proposed to reconstruct the 3D point clouds (PCs) of specific brains from one single incomplete image with low latency . 3D shape reconstruction is essential in the navigation of minimally-invasive and auto robot-guided surgeries whose operating environments are indirect and narrow .…

Conditional computation and modular networks have been recently proposed formultitask learning and other problems as a way to decompose problem solving into multiple reusable computational blocks . We propose a new approach forlearning modular networks based on the isometric version of ResNet with allresidual blocks having the same configuration and the same number of parameters .…

The method significantly compressesneural network weights to a sparse ternary of [-1,0,1 . It can compress aResNet-18 model from 46 MB to 955KB and a ResNet-50 model from 99 MB to 3.3MB (~30x) The top-1 accuracy on ImageNet drops slightly from 69.7% to65.3% and from 76.15% to 74.47% .…

DA-VSN is a domain adaptive video segmentation network that addresses domain gaps in videos by temporal consistency regularization (TCR) for consecutive frames of target-domain videos . The network is based on cross-domain TCR that guides the prediction of target frames to have similar temporal consistency as that of source frames (learnt from annotated source data) viaadversarial learning .…

The full range of activity in a temporal network is captured in its edgeactivity data — time series encoding the tie strengths or on-off dynamics of each edge in the network . In many practical applications, edge-leveldata are unavailable, and the network analyses must rely instead on nodeactivity data .…

Scene text editing (STE) is a challenging task due to a complex intervention between text and style . We propose a novel representational learning-based STE model that employs textual information as well as visual information . Our experiments demonstrate that RewriteNet achieves betterquantitative and qualitative performance than other comparisons .…

Unrealistic feature suppression (UFS) module keeps high-quality features and suppresses unrealistic features . UFS module keeps the training stability of networks and improves the quality ofgenerated images . We demonstrate effectiveness of the module on various models such as WGAN-GP, SNGAN, and BigGAN .…

Feature pyramids and iterative refinement have recently led to great progress in optical flow estimation . However, downsampling in feature pyramids can cause blending of foreground objects with background . We propose anovel Residual Feature Pyramid Module (RFPM) which retains important details inthe feature map without changing the overall design of the overall iterative .…

Skyline plays a pivotal role in mountainous visual geo-localization and localization/navigation of planetary rovers/UAVs and virtual/augmented realityapplications . We present a novel mountainous skyline detection approach wherewe adapt a shallow learning approach to learn a set of filters to discriminate between edges belonging to sky-mountain boundary and others coming from different regions .…

Single-photon cameras (SPCs) are an emergingsensing modality that are capable of capturing images with high sensitivity . Despite having minimal read-noise, images captured by SPCs in photon-starved conditions still suffer from strong shot noise . We propose photon scale-space a collection of high-SNR imagesspanning a wide range of photons-per-pixel (PPP) levels as guides to train inference model on low photon flux images .…

Deep generative models have made great progress in synthesizing images with human poses and transferring poses of one person to others . Most existing approaches explicitly leverage the pose information extracted from the source images as a conditional input for the generative networks .…