Transfer Learning for Human & AI

Network neutrality is related to the non-discriminatory treatment of packets on the Internet . Any deliberate discrimination of traffic of one application while favoring others violates the principle of neutrality . Many countries haveenforced laws against such discrimination . To enforce such laws, one requirestools to detect any net neutrality violations .…

Weighted voting is a conventional approach to improving the performance of systems based on commonly-used majority quorum systems in heterogeneous environments . In long-lived systems, a weight reassignment protocol is required to reassign weights over time in order to accommodate performance variations accordingly .…

We study combinatorial inequalities for various classes of set systems . We use the language formulation of greedoids which allows a linear algebraic setup . The underlying non-commutative nature of matrices associated withgreedoids allows us to proceed beyond polymatroids and prove the equalityconditions .…

Symbolic Interactive Language Grounding benchmark (SILG) unifies a collection of diverse grounded language learning environments . SILG consists of grid-world environments that require generalization to new dynamics, entities, and partially observed worlds (RTFM,Messenger, NetHack) and symbolic counterparts of visual worlds that require interpreting rich natural language with respect to complex scenes (ALFWorld, Touchdown) Together, these environments provide diverse groundingchallenges in richness of observation space, action space, languagespecification, and plan complexity .…

The field of style transfer (ST) has been growing rapidly, it has been hampered by a lack of standardized practices for automatic evaluation . Unlike previous evaluations, which focus solely on English, we expand our focus to Brazilian-Portuguese, French, and Italian .…

Python has become a dominant programming language for emerging areas likeMachine Learning (ML), Deep Learning (DL), and Data Science (DS) An attractive feature of Python is that it provides easy-to-use programming interface while allowing library developers to enhance performance of their applications .…

An emerging problem in themodern era is fake news detection — many easily available pieces of information are not necessarily factually correct . In this work we explore how differentdocument representations, ranging from simple symbolic bag-of-words, to contextual, neural language model-based ones can be used for efficient fakenews identification .…

Work aims to connect the Automotive User Interfaces (Auto-UI) and Conversational User . Interfaces communities through discussion of their view of the future of automotive . user interfaces . The workshop aims to encourage creative consideration of optimistic and pessimistic .…

In this work, we introduce the pattern-domain pilot design paradigm based on a “superposition of orthogonal-building-blocks” with significantly largercontention space to enhance the massive machine-type communications (mMTC)random access (RA) performance in massive multiple-input multiple-output (MIMO) systems . The increased contention space, i.e.,…

The recent phenomenal success of language models has reinvigorated machinelearning research . One important problem class that has remained elusive however is purposeful adaptive behavior . Currently there is a common perception that sequence models “lack the understanding of the causeand effect of their actions” leading them to draw incorrect inferences due toauto-suggestive delusions .…

Nonlinear entropy stability analysis is used to derive entropy stable no-slipwall boundary conditions for the Eulerian model proposed by Sv\”{a}rd (PhysicaA: Statistical Mechanics and its Applications, 2018) and its spatialdiscretization based on entropy stable collocated discontinuous Galerkinoperators with the summation-by-parts property for unstructured grids .…

Proactive tile-based virtual reality (VR) video streaming can use the traceof FoV and eye movement to predict future requested tiles, then renders and delivers the predicted tiles before playback . Quality of experience (QoE) depends on the combined effect of tile prediction and consumed resources .…

The primordial power spectrum informs the possible inflationary histories ofour universe . Given a power spectrum, the ensuing cosmic microwave background is calculated and compared to the observed one . We show that the validity of analyticapproximations of the scalar index roughly respects the large-field/small-field dichotomy .…

Normal estimation on 3D point clouds is a fundamental problem in 3D vision and graphics . Current methods often show limited accuracy in predicting normals at sharp features (e.g., edges and corners) and less robustness to noise . We are motivated that local patches on isotropic and anisotropic surfaces have similar or distinct normals .…

Age of Information (AoI) reflects the time that is elapsed from the generation of a packet by a 5G user equipment(UE) to the reception of the network . A design of an AoI-aware radio resource scheduler forUEs via reinforcement learning is proposed in this paper .…

We present a generalised phase field-based formulation for predicting fatiguecrack growth in metals . The theoretical framework aims at covering a wide range of material behaviour . A monolithic solution scheme based onquasi-Newton algorithms is presented and shown to significantly outperformstaggered approaches .…

We describe the 2021 Key Point Analysis (KPA-2021) shared task on key pointanalysis that we organized as a part of the 8th Workshop on Argument Mining(ArgMining 2021) at EMNLP 2021 . We outline various approaches and discuss the results of the shared task .…

Self-attention models pre-trained on large amounts of unannotated data have been hugely successful when fine-tuned on downstream tasks from a variety of domains and languages . We build a singleencoder with the BERT objective on unlabeled text together with the w2v-BERT objective .…

Large-scale pretrained language models have been shown to learneffective linguistic representations for many NLP tasks . We propose a simple approach to incorporate speaker social context into the learnedrepresentations of large-scale language models . We evaluate our approach on geographically-sensitivelanguage-modeling tasks and show a substantial improvement (more than 100% relative lift on MRR) compared to baselines .…

ComputationalGraph Completion (CGC) problem addressed by the proposed framework could be interpreted as a generalization of that of solving linearsystems of equations to that of approximating unknown variables and functions with noisy, incomplete, and nonlinear dependencies . The proposed framework can be used to recover unknown functions with much scarcer data by exploitinginterdependencies between multiple functions and variables .…

Colosseum is an open-access and publicly-available large-scale wirelesstestbed for experimental research via virtualized and softwarized waveforms and protocol stacks on a fully programmable, “white-box” platform . Through 256 state-of-the-art Software-defined Radios and a Massive Channel Emulator core, the system can model virtually any scenario, enabling the design, developmentand testing of solutions at scale in a variety of deployments and channelconditions .…

We develop methodology for three-dimensional (3D) radial visualization of multidimensional datasets . We show that this uniform distribution provides the best visualization with minimal artificial visualcorrelation for data with uncorrelated variables . We use RadViz3D to illustrate (i) the chemical composition of Longquan celadon ceramics and their Jingdezhen imitation overcenturies, and (ii) US regional SARS-Cov-2 variants’ prevalence in the Covid-19pandemic during the summer 2021 surge of the Delta variant .…

In this paper, we propose a $C^{0$ interior penalty method for $m$th-Laplaceequation on bounded Lipschitz polyhedral domain in $\mathbb{R}^{d}$ The standard $H^{1$-conforming piecewise$r$-th order polynomial space is used to approximate the exact solution $u$where $r$ can be any integer greater than or equal to $m$.…

We introduce a finite volume scheme to solve isotropic 3-wave kineticequations . Our numerical energy cascade rates are in good agreement with the theoretical one obtained by Soffer and Tran . Our finite volume algorithm relies on a new identity, that allows one to reduce the number of terms needed to be approximated in the collision operators .…

With autonomous vehicles (AVs) a major concern is the inability to give quantitative assurance of safety . We demonstrate an approach to achieving more moderate,but useful, confidence . This formalises mathematically the approach of operating a system on a limited basis in the hope that Mishap-free operation will confirm one’s confidence in its safety and allow more extensive operation: a process of “bootstrapping” of confidence .…

The study demonstrates the feasibility of creating a federated NLP algorithmdevelopment and benchmarking platform to enhance multi-institution clinical NLP study . The generated corpora derived out of thetexts from multiple intuitions and gold standard annotated are tested on asingle institution’s rule set .…

Asynchronous iterations are more and more investigated for both scaling and fault-resilience purpose on high performance computing platforms . This paper advocates a novel application direction targeting time-decomposed time-parallel approaches . It turned out that Parareal andasync-Parareal feature very close convergence conditions, asymptoticallyequivalent, including the finite-time termination property .…

An emerging problem in themodern era is fake news detection — many easily available pieces of information are not necessarily factually correct . In this work we explore how differentdocument representations, ranging from simple symbolic bag-of-words, to contextual, neural language model-based ones can be used for efficient fakenews identification .…

The task of talking head generation is to synthesize a lip synchronizedtalking head video by inputting an arbitrary face image and audio clips . Most existing methods ignore the local driving information of the mouth muscles . In this paper, we propose a novel recurrent generative network that uses bothaudio and speech-related facial action units as the driving information .…

Requirements driven search-based testing (also known as falsification) has proven to be a practical and effective method for discovering erroneous behaviors in Cyber-Physical Systems . The absence of finite time guarantees is a majorlimitation which prevents falsification methods from being utilized incertification procedures .…

Can every connected graph burn in $\lceil \sqrt{n} \rceil $ steps? While this remains open, we prove that it is asymptotically true when the graphis much larger than its growth . We prove that theconjecture for graphs of bounded growth boils down to a finite number of cases .…

In this paper, we study a matching market model on a bipartite network where agents on each side arrive and depart stochastically by a Poisson process . We design a mechanism that decides not only which agentsto match, but also when to match them, to minimize the expected number ofunmatched agents .…

Symbolic Interactive Language Grounding benchmark (SILG) unifies a collection of diverse grounded language learning environments . SILG consists of grid-world environments that require generalization to new dynamics, entities, and partially observed worlds (RTFM,Messenger, NetHack) and symbolic counterparts of visual worlds that require interpreting rich natural language with respect to complex scenes (ALFWorld, Touchdown) Together, these environments provide diverse groundingchallenges in richness of observation space, action space, languagespecification, and plan complexity .…

In this paper we illustrate that paraorthogonality on the unit circle$\mathbb{T}$ is the counterpart to orthog on $mathbb {R}$ when we are interested in the spectral properties . We characterize quasi-paraorthogonalpolynomials on theunit circle as the analogues of the quasi-orthosomials on $mathbb R}$.…

Large-scale pretrained language models have been shown to learneffective linguistic representations for many NLP tasks . We propose a simple approach to incorporate speaker social context into the learnedrepresentations of large-scale language models . We evaluate our approach on geographically-sensitivelanguage-modeling tasks and show a substantial improvement (more than 100% relative lift on MRR) compared to baselines .…

We study the role of local effects and finite size effects in reachingcoordination and in equilibrium selection in different types of two-player coordination games . We investigate three update rules — the replicatordynamics (RD), the best response (BR), and the unconditional imitation (UI) –for coordination games on random graphs .…

In reinforcement learning, pre-trained low-level skills have the potential to facilitate exploration . However, prior knowledge of the downstream task is required to strike the right balance between generality (fine-grained control) and specificity (faster learning) in skill design . We alleviate the need for prior knowledge by proposing a .…

Many-Fermion Dynamics-nuclear, or MFDn, is a configuration interaction (CI)code for nuclear structure calculations . It is a platform-independent Fortran90 code using a hybrid MPI+X programming model . For CPU platforms the application has a robust and optimized OpenMP implementation for shared memoryparallelism.…

Distributed energy resources (DERs) are growing rapidly and are reshaping power systems . Feed-in-tariff (FIT) is usually adopted by utilities to pay DER owners certain rates for supplying energy to the grid . An alternative to FIT is amarket-based approach; that is, consumers trade energy in anauction-based peer-to-peer (P2P) market .…

Reinforcement learning is traditionally used in a tabula rasasetting and must explore and learn everything from scratch . We show that this strategy can be computed atlittle cost, by solving a convex optimization problem . By replacing the usualepsilon-greedy strategy in a DQN, we demonstrate its potential in several environments with various dynamic structures .…

Normal estimation on 3D point clouds is a fundamental problem in 3D vision and graphics . Current methods often show limited accuracy in predicting normals at sharp features (e.g., edges and corners) and less robustness to noise . We are motivated that local patches on isotropic and anisotropic surfaces have similar or distinct normals .…

Socialbots are software-driven user accounts on social platforms, actingautonomously (mimicking human behavior) with the aims to influence theopinions of other users or spread targeted misinformation for particular goals . As such, there have been several computational efforts toauto-detect the socialbots .…

We adapt a continuous data assimilation scheme to the case of moving observers for the 2Dincompressible Navier-Stokes equations . We propose and test computationallyseveral movement patterns (which we refer to as “the bleeps, the sweeps and thecreeps) We end with a discussion of possible applications to real-world datacollection strategies that may lead to substantial improvements in predictivecapabilities .…

Automated market makers (AMMs) are smart contracts that automatically trade electronic assets according to a mathematical formula . This paper investigates how an AMM’s formula affects the interests of liquidity providers, who endow the AMM with assets, and traders, who exchange one asset for another at theAMM’s rates .…

Multi-agent reinforcement learning (MARL) algorithms have made promisingprogress in recent years . But existing MARL algorithms still suffer from the sample inefficiency problem . State-based episodic memory (SEM) is the first work to introduce EM into MARL . It has lower space complexity and time complexity than state and action based EM (SAEM), which is originally proposed for single-agent learning .…

Probabilistic numerical solvers for ordinary differential equations computeposterior distributions over the solution of an initial value problem viaBayesian inference . Weshow that second-order differential equations should be directly provided to the solver, instead of transforming the problem to first order .…

Projection-based reduced order models are effective at approximatingparameter-dependent differential equations that are parametrically separable . When parametric separability is not satisfied, which occurs in both linear andnonlinear problems, projection-based methods fail to adequately reduce the complexity of these problems . Devising alternative models is crucial for obtaining efficient and accurate approximations to expensive high-fidelity models, such as those by a neural-network .…

We present a novel characterization of the mapping of multiple parallelismforms onto hierarchical accelerator system . We experimentally verify the substantial effect of these mappings onall-reduce performance (up to 448x) We offer a novel syntax-guided programsynthesis framework that is able to decompose reductions over one or moreparallelism axes to sequences of collectives in a hierarchy- and mapping-awareway .…

The multi-headmulti-modal DIN improves the recommendation prediction effect, and outperforms state-of-the-art methods on various comprehensive indicators . Researchers have optimized more on therecommendation model network structure, they have less research on enrichingrecommendation models, and there is still room for in-depth model optimization .…

Requirements driven search-based testing (also known as falsification) has proven to be a practical and effective method for discovering erroneous behaviors in Cyber-Physical Systems . The absence of finite time guarantees is a majorlimitation which prevents falsification methods from being utilized incertification procedures .…