Transfer Learning for Human & AI

The shift-enabled property of an underlying graph is essential in designing distributed filters . In particular, popular graph models ER, WS, BA random graph are used, weighted and unweighted, as well as signed graphs . The behaviour is not observed for weighted connected graphs, but very dense graphs are not shifted-enabled .…

We propose a framework to study the effect of local recovery requirements of codeword symbols on the dimension of linear codes . The locality constraints of a linearcode are stipulated by a matrix $H$ of $H $ of $0$’s and $H’s .…

The goal of the current work is to explore direction-dependent damageinitiation and propagation within an arbitrary anisotropic solid . The results of the latter models are shown to be insensitive withrespect to the length scale parameter for the isotropic case .…

We propose an algorithm for identifying clusters in the Markov chain associated with $T$ . The algorithm isrecursive in nature, and in order to identify clusters, it uses the signpattern of a left singular vector associated with the second smallest singularvalue of the Laplacian matrix $I-T.$…

We propose a novel framework for predicting the factuality of reporting of news media outlets by studying user attention cycles in their YouTubechannels . We design a rich set of features derived from thetemporal evolution of the number of views, likes, dislikes, and comments for avideo, which we then aggregate to the channel level .…

As online social platforms increasingly impact society, scholars andactivists have criticized the centralized and unaccountable nature of many of their governance structures . Some have proposed governance that represents a broader range of stakeholders through the creation of multiplecenters of power .…

PROMELA (Process Meta Language) is a high-level specification languagedesigned for modeling interactions in distributed systems . It is used as an input language for the model checker SPIN (Simple Promela INterpreter) The main characteristics are non-determinism, process communication through synchronous as well as asynchronous channels .…

Fault diagnosis of rotating machinery is an important engineering problem . Fault diagnosis methods based on Convolutional Neural Network(CNN) and Recurrent Neural Network (RNN) have been mature, but Transformer has not been widely used in the field of fault diagnosis .…

The limited representation of minorities and disadvantaged populations in large-scale clinical and genomics research has become a barrier to translating precision medicine research into practice . Risk prediction models are often found to be underperformed inthese underrepresented populations . In this paper, we propose a two-way data integration strategy that integrates heterogeneous data from diverse populations and from multiple healthcare institutions via a federated transfer learning approach .…

A long-standing conjecture by Heath, Pemmaraju, and Trenk states that theupward book thickness of outerplanar DAGs is bounded above by a constant . We show that the conjecture holds for subfamilies of upwardouterplanar graphs, namely those whose underlying graph is aninternally-triangulated outerpath or a cactus .…

Stylized image captioning systems aim to generate a caption not onlysemantically related to a given image but also consistent with a given styled description . One of the biggest challenges with this task is the lack ofsufficient paired stylized data .…

Spatial telecommunication systems have evolved along the years, leading to some concerns that telecommunication companies are facing today . The maininquietude is the ability to provide quality service to customers or users in adense regime . What is the best possibleconfigurations of base stations and users that maximizes quality service?…

Congestion pricing means that the operator increases prices at the time of congestion and the traffic demand are expected to decrease . Shadow prices are an optimal way of congestion pricing: users are chargedshadow prices, i.e., the expectations of future losses because of blockedconnections .…

State-of-the-art deep-learning compilers like TVM and Halide incorporate alearning-based performance model to search the space of valid implementation of a given deep learning algorithm . For a given application, the modelgenerates a performance metric such as the run time without executing the application on hardware .…

Developers rely on build systems to generate software from code . Simple build specifications are easy to write, but limit incremental work . We introduce LaForge, a novel build tool that eliminates the need to specifydependencies or incremental build steps .…

Predictive models for clinical outcomes that are accurate on average in apatient population may underperform drastically for some subpopulations . No approach performs better, for each patient subpopulation examined, than standardlearning procedures using the entire training dataset . These results imply that when it is of interest to improve model performance, it may be necessary to doso via techniques that implicitly or explicitly increase the effective samplesize, such as increasing the number of patient sub-populations that are of interest .…

Recently, many datasets have been proposed to test the systematicgeneralization ability of neural networks . The companion baseline Transformers, typically trained with default hyper-parameters from standard tasks, are shownto fail dramatically . Here we demonstrate that by revisiting modelconfigurations as basic as scaling of embeddings, early stopping, relativepositional embedding, and Universal Transformer variants, we can drasticallyimprove the performance of Transformers on systematic generalization .…

Evolutionary computation is one of the main applications in the field of evolutionary computation . With programsynthesis techniques, software developers could be supported in theireveryday work . Even users without any programming knowledge could be empowered to automate repetitive tasks and implement their own newfunctionality .…

Chv\'{a}tal and Klincsek (1980) gave an $O(n^3)$-time algorithm for the problem of finding a maximum-cardinality convex subset of an arbitrary givenset $P$ of $n$ points in the plane . We show the problem is NP-hard when $k$ is an arbitrary input parameter .…

Chv\'{a}tal and Klincsek (1980) gave an $O(n^3)$-time algorithm for the problem of finding a maximum-cardinality convex subset of an arbitrary givenset $P$ of $n$ points in the plane . We show the problem is NP-hard when $k$ is an arbitrary input parameter .…

Chv\'{a}tal and Klincsek (1980) gave an $O(n^3)$-time algorithm for the problem of finding a maximum-cardinality convex subset of an arbitrary givenset $P$ of $n$ points in the plane . We show the problem is NP-hard when $k$ is an arbitrary input parameter .…

Fault localization is one of the most time-consuming and error-prone parts of software debugging . The tool employsSpectrum-based fault localization (SBFL) to help Python developers analyze their programs and generate useful data at run-time to beused, then to produce a ranked list of potentially faulty program elements .…

A 3-SAT problem is called positive and planar if all the literals are positive and the clause-variable incidence graph (i.e., SAT graph) is planar . In this paper we prove that a positive planar NAE3SAT is always satisfiable when the underlying SAT graph is 3-connected, and asatisfiable assignment can be obtained in linear time .…

Brushless DirectCurrent motor (BLDC motor) uses electronic closed-loop controllers to switch DCcurrent to the motor windings and produces the magnetic fields . The motor is modeled in the MATLAB/Simulink and the speedcontrol is obtained with a PI controller . The acquired data is then fed into binary artificial neuralnetworks and as a result, the ANN model predicts the corresponding parameters close to the simulation results .…

Graph filters are an emerging paradigm that systematizes informationpropagation in graphs as transformation of prior node values, called graphsignals, to posterior scores . In this work, we study the problem of mitigatingdisparate impact, i.e. posterior score differences between a protected set ofsensitive nodes and the rest, while minimally editing scores to preserverecommendation quality .…

Historians and researchers trust web archives to preserve social mediacontent that no longer exists on the live web . Twitter forced the use of its new UI in June 2020 . Most web archives were unable to archive the new UI, resulting in archived Twitter pages displaying Twitter’s “Something went wrong” error .…

This paper is an exploratory study of two epidemiological questions on aworldwide basis . How fast is the disease spreading? Are the restrictions(especially mobility restrictions) for people bring the expected effect? To answer the first question, we propose a tool for estimating the reproductionnumber of epidemic (the number of secondary infections $R_t$) based on the SEIR model .…

The goal of this article is to introduce the Hamiltonian Monte Carlo (HMC)method . We focus on the “idealized” case, where onecan compute continuous trajectories exactly . We show that idealized HMCpreserves $pi$ and we establish its convergence when $f$ is strongly convexand smooth .…

Stochastic computing (SC) is a low-cost computational paradigm that has promising applications in digital filter design, image processing and neuralnetworks . Fundamental to these applications is the weighted addition operation which is most often implemented by a multiplexer (mux) tree .…

A 3-SAT problem is called positive and planar if all the literals are positive and the clause-variable incidence graph (i.e., SAT graph) is planar . In this paper we prove that a positive planar NAE3SAT is always satisfiable when the underlying SAT graph is 3-connected, and asatisfiable assignment can be obtained in linear time .…

Twitter data has become established as a valuable source of data for various applications . For many applications, it is necessary to know where Twitter posts (tweets) were sent from or what locationthey refer to . Researchers have frequently used exact coordinates provided in asmall percentage of tweets, but Twitter removed the option to share these coordinates in mid-2019 .…

This is the first large-scale study of DH research that involved a huge number of bibliographic data sources, including open and proprietary databases . Citations from/to DH publications showed relations between DH and research in Computer Science, Linguistics, Psychology, and Pedagogical & Educational Research .…

Freight carriers rely on tactical plans to satisfy demand in a cost-effective way . The periodic demand is the demand that is expected torepeat in each period of the planning horizon . Laage et al. (2021) recently introduced the Periodic Demand Estimation (PDE)problem and showed that it has a high value .…

Facial detection and analysis systems have been deployed by large companies and critiqued by scholars and activists for the past decade . We present the first of its kind detailedbenchmark of the robustness of three such systems: Amazon Rekognition,Microsoft Azure, and Google Cloud Platform .…

The animation community has spent significant effort trying to ease riggingprocedures . The increasing availability of 3D datamakes manual rigging infeasible . However, object animations involve understanding elaborate geometry and dynamics . This paper proposes a functional approach for skeletontransfer that uses limited information and does not require a complete match between the geometries .…

We consider the problem of drawing an outerplanar graph with $n$ vertices with at most one bend per edge if the outer face is already drawn as a simplepolygon . We prove that it can be decided in $O(nm)$ time if such a drawingexists, where $m\le n-3$ is the number of interior edges .…

We develop an adaptive group testing algorithm that uses the proposed set formation method recursively . We prove novel upper bounds on thenumber of tests for the proposed algorithm . We show thatour algorithm outperforms the state of the art, and performs close to theentropy lower bound.…

When a sufficient amount of within-domain text may not be available, expanding a seed corpus of relevant documents from large-scale web data poses several challenges . The authors propose a novel truncated sparse document bit-vectorrepresentation, termed Signature Assisted Unsupervised Corpus Expansion(SAUCE) The SAUCE can reduce the computational burden while ensuring high within-Domain lexical coverage, especially under limited seed corpora scenarios.…

We present a new quantum algorithm for estimating the mean of a real-valued random variable obtained as the output of a quantum computation . Our estimatorachieves a nearly-optimal quadratic speedup over the number of classical i.i.d.samples needed . We obtain new quantum algorithms for the .…

Passage retrieval and ranking is a key task in open-domain question answering and information retrieval . Current effective approaches mostly rely on deep language model-based retrievers and rankers . Authors propose a simple typos-aware training framework for DR and BERT re-ranker to address this issue .…

This paper presents Macquarie University’s participation to the BioASQSynergy Task, and BioasQ9b Phase B . The Synergy Task is an end-to-end questionanswering task on COVID-19 . Our system splitthe snippets into candidate sentences and used BERT variants under a sentenceclassification setup .…

Freight carriers rely on tactical plans to satisfy demand in a cost-effective way . The periodic demand is the demand that is expected torepeat in each period of the planning horizon . Laage et al. (2021) recently introduced the Periodic Demand Estimation (PDE)problem and showed that it has a high value .…

A 3-SAT problem is called positive and planar if all the literals are positive and the clause-variable incidence graph (i.e., SAT graph) is planar . In this paper we prove that a positive planar NAE3SAT is always satisfiable when the underlying SAT graph is 3-connected, and asatisfiable assignment can be obtained in linear time .…

Recommender systems typically operate on high-dimensional sparse user-item matrix matrix . We propose a Global-Local Kernel-based matrix completionframework, named GLocal-K . Our model outperforms the state-of-the-artbaselines on three collaborative filtering benchmarks: ML-100K, ML-1M, andDouban. We apply our model under the extreme low-resource setting, which includes only a user item rating matrix, with no side information, to an extreme low resource setting .…

Software Engineering (SE) researchers extensively perform experiments with human subjects . Well-defined samples are required to ensure validity . Samples are selected by either by purpose or by convenience, limiting the generalizability of results . We outline a new selectionmethodology, by defining the characteristics of the desired population, .…

A 3-SAT problem is called positive and planar if all the literals are positive and the clause-variable incidence graph (i.e., SAT graph) is planar . In this paper we prove that a positive planar NAE3SAT is always satisfiable when the underlying SAT graph is 3-connected, and asatisfiable assignment can be obtained in linear time .…

We present a new quantum algorithm for estimating the mean of a real-valued random variable obtained as the output of a quantum computation . Our estimatorachieves a nearly-optimal quadratic speedup over the number of classical i.i.d.samples needed . We obtain new quantum algorithms for the .…

Prover-Verifier Games (PVGs) is a game-theoretic framework to encourage learningagents to solve decision problems in a verifiable manner . The PVG consists of two learners with competing objectives: a trusted verifier network tries tochoose the correct answer, and a more powerful but untrusted prover network attempts to persuade the verifier of a particular answer .…

We present a new quantum algorithm for estimating the mean of a real-valued random variable obtained as the output of a quantum computation . Our estimatorachieves a nearly-optimal quadratic speedup over the number of classical i.i.d.samples needed . We obtain new quantum algorithms for the .…

We present a new quantum algorithm for estimating the mean of a real-valued random variable obtained as the output of a quantum computation . Our estimatorachieves a nearly-optimal quadratic speedup over the number of classical i.i.d.samples needed . We obtain new quantum algorithms for the .…