Overprotective Training Environments Fall Short at Testing Time Let Models Contribute to Their Own Training

Despite important progress, conversational systems often generate dialoguesthat sound unnatural to humans . We conjecture that the reason lies in their training and testing conditions . During training, they learn to generate utterance given the human dialogue history . On the other hand, duringtesting, they must interact with each other, and hence deal with noisy data .…

Dependency Graph to String Statistical Machine Translation

We present graph-based translation models which translate source graphs intotarget strings . Source graphs are constructed from dependency trees with extralinks so that non-syntactic phrases are connected . We provide two implementations of the model with different restrictions so that source graphs can be parsed efficiently .…

Leveraging Unlabeled Data for Entity Relation Extraction through Probabilistic Constraint Satisfaction

We study the problem of entity-relation extraction in the presence ofsymbolic domain knowledge . Such knowledge takes the form of an ontology defining relations and their permissible arguments . Previous approaches set out to integrate such knowledge in their learning approaches either through self-training, or through approximations that lose the precise meaning of the logical expressions .…

On the Impossibility of Post Quantum Black Box Zero Knowledge in Constant Rounds

We investigate the existence of constant-round post-quantum black-box zero-knowledge protocols for $\mathbf{NP$ . Main result points out a fundamental difference between post-Quantum and classical zero-knowledges . We conclude that unless we use non-black-box techniques or relax certain security requirements, the protocols for$\mathf{NP}$ exist ifand only if we usenon-black box techniques .…

An Efficient Simulation of Quantum Secret Sharing

In quantum cryptography, quantum secret sharing $(QSS)$ is a fundamentalprimitive . $QSS$ can be used to create complex and secure multiparty quantumprotocols . Here, we propose a secure $d$-level protocol for sharing a secret withefficient simulation . This protocol is more secure, flexible, and practical ascompared to the existing protocols: $(n, n)$ threshold $2$ level and $(t, n)’s threshold $D$ level with a trusted player, where $n$denotes the number of players and $t$ denotes the threshold number of .…

Encrypted Value Iteration and Temporal Difference Learning over Leveled Homomorphic Encryption

We consider an architecture of confidential cloud-based control synthesis based on Homomorphic Encryption (HE) Our study is motivated by the recentsurge of data-driven control such as deep reinforcement learning . We first show that the impact of the encryption noise under the Cheon-Kim-Kim Kim-Song (CKKS) encryption scheme on the convergence of themodel-based tabular Value Iteration (VI) can be analytically bounded .…

Tubu io Decentralized Application Development Test Workbench

Decentralized services are increasingly being developed and their proper usage in different areas is being experimented with . Autonomous codes, whichare also called smart contracts, can be developed with Integrated DevelopmentEnvironments (IDE) However, these tools lack live environment tests . Tubu-io decentralizedapplication development workbench is developed to serve as an efficient way for programmers to deploy smart contracts on the blockchain networks and interact with them easily .…

Joint Resource Allocation and Cache Placement for Location Aware Multi User Mobile Edge Computing

Mobile edge computing (MEC) has emerged as apromising technique to reinforce the computation capability of the resource-constrained mobile devices . To exploit the cloud-like functions at thenetwork edge, service caching has been implemented to (partially) reuse thecomputation tasks, thus effectively reducing the delay incurred by dataretransmissions and/or the computation burden due to repeated execution of the same task .…

Multi sequence Spreading Random Access MSRA for Compressive Sensing based Grant free Communication

Multi-sequence spreading random access (MSRA) employs multiple spreading sequences to spread the different symbols of a user . MSRA provides code diversity, enabling the multi-user detection (MUD) to be modeled into a well-conditioned multiple measurement vector (MMV) CS problem . The theoretical analysis showsthat with MSRA activity misdetection falls exponentially while the size ofGF-RA frame is increased .…

Hierarchical Hybrid Error Correction for Time Sensitive Devices at the Edge

Computational storage, known as a solution to significantly reduce thelatency by moving data-processing down to the data storage, has received wideattention because of its potential to accelerate data-driven devices at the edge . To meet the insatiable appetite for complicated functionalities tailored for intelligent devices such as autonomous vehicles, properties includingheterogeneity, scalability, and flexibility are becoming increasinglyimportant .…

Forward and Backward Bellman equations improve the efficiency of EM algorithm for DEC POMDP

Decentralized Partially Observable Markov Decision Process (DEC-POMDP) modelssequential decision making problems by a team of agents . In EM, the forward-backward algorithm needs to be calculated up to the infinite horizon, which impairs the computational efficiency . In this paper, we propose Bellman EM algorithm (BEM) and Modified BellmanEM algorithm (MBEM) by introducing the forward and backward Bellman equations into EM .…

An ADMM Newton CNN Numerical Approach to a TV Model for Identifying Discontinuous Diffusion Coefficients in Elliptic Equations Convex Case with Gradient Observations

Models with total variational (TV)regularization have been widely studied for this problem . Theoretically required nonsmoothness property of the TV regularization and the hidden convexity of the models are usually sacrificed when numerical schemes are considered in the literature . We show that ADMM-Newton-CNN approach is easily implementable and very efficient even for higher-dimensional spaces with fine mesh discretization .…

A note on sampling recovery of multivariate functions in the uniform norm

We study the recovery of multivariate functions from reproducing kernelHilbert spaces in the uniform norm . Our main interest is to obtainpreasymptotic estimates for the corresponding sampling numbers . We obtain results in terms of the decay of related singular numbers of the compactembedding into $L_2(D,\varrho_D)$ multiplied with the supremum of theChristoffel function of the subspace spanned by the first $m$ singularfunctions .…

The Multiscale Perturbation Method for Two Phase Reservoir Flow Problems

The MPM-2P is based on domain decomposition and combinesthe Multiscale Perturbation Method (MPM) with the Multiscales Robin CoupledMethod (MRCM) It is a new procedure for the fast, accurate andnaturally parallelizable numerical solution of two-phase, incompressible,immiscible displacement in porous media . Hundreds of MRCM solutions can be replaced byinexpensive MPM2P solutions, and water breakthrough can be simulated with veryfew updates of the MRCm set of multiscale basis functions .…

Train Deep Neural Networks in 40 D Subspaces

DynamicLinear Dimensionality Reduction (DLDR) dramatically reduces theparameter space to a variable subspace of significantly lower dimension . The experimental results strongly support the dimensionalityreduction performance . For many standard neural networks, optimizing over only40 variables, one can achieve comparable performance against the regulartraining over thousands or even millions of parameters, says the authors .…

Quality Evolvability ES Evolving Individuals With a Distribution of Well Performing and Diverse Offspring

Evolvability algorithms aim to automatically learn good genetic representations . They aim to find a single individual with a diverse and well-performing distribution of offspring . By doing so QualityEvolvability is forced to discover more evolvable representations . It can learn faster than objective-based methods and can handle deceptive problems, say authors .…

Most Efficient Sensor Network Protocol for a Permanent Natural Disaster Monitoring System

The adaptive sensornode management protocol (ASMP) makes system componentsto systematically control their performance to conserve the energy . Through this method, ASMP achieves both energyconservation and service quality . For theoptimized environment sampling, we proposed the adaptive sampling algorithm formonitoring (ASA-m) We propose a novel energy management protocol forenergy harvesting wireless sensor networks (EH-WSNs), named the Adaptation Sensor Network .…

UAV Communications for Sustainable Federated Learning

Federated learning (FL), invented by Google in 2016, has become a hot research trend . enabling FL in wireless networks has to overcome the limited battery challenge of mobile users . In this regard, we propose to applyunmanned aerial vehicle (UAV)-empowered wireless power transfer to enables sustainable FL-based wireless networks .…

External Forces Resilient Safe Motion Planning for Quadrotor

This paper proposes a systematic (re)planning framework that can resiliently generate safe trajectories under volatile conditions . It guarantees collision-free by constricting the ellipsoid of the quadrotor body expanded with the forwardreachable sets (FRSs) within safe convex polytopes . Our method is validated insimulations and real-world experiments with different sources of externalforces .…

Learning Continuous Cost to Go Functions for Non holonomic Systems

This paper presents a supervised learning method to generate continuous cost-to-go functions of non-holonomic systems directly from the workspaced description . Supervision from informative examples reduces training time andimproves network performance . The manifold representing the optimaltrajectories of a non-Holonomic system has high-curvature regions which can not be efficiently captured with uniform sampling .…

The Visual Inertial Dynamical UAV Dataset

The VID dataset contains hard synchronized imageryand inertial measurements, with accurate ground truth trajectories forevaluating common visual-inertial estimators . The proposed dataethighlights the measurements of rotor speed and motor current, dynamical inputs,and ground truth 6-axis force data to evaluate external force estimation .…

Unsupervised Feature Learning for Manipulation with Contrastive Domain Randomization

Robotic tasks such as manipulation with visual inputs require image featuresthat capture the physical properties of the scene . Recently, it has been suggested to learn such features in an unsupervised manner from simulated, self-supervised, robotinteraction . Torobustify the simulation-to-real transfer, domain randomization (DR) wassuggested for learning features that are invariant to irrelevant visualproperties such as textures or lighting .…

Force Sensing in Robot assisted Keyhole Endoscopy A Systematic Survey

Instrument-tissue interaction forces in Minimally Invasive Surgery (MIS) provide valuable information that can be used to provide haptic perception, monitor tissue trauma, develop training guidelines, and evaluate the skilllevel of novice and expert surgeons . This article provides a comprehensive systematic review of the current forcesensing research aimed at RAS and, more generally, keyhole endoscopy, in whichinstruments enter the body through small incisions .…

An Efficient Calibration Method for Triaxial Gyroscope

This paper presents an efficient servomotor-aided calibration method for thetriaxial gyroscope . The entire calibration process only takes about one minute,and high-precision equipment is not used . The calibration results of the proposed method areverified by comparing with a traditional turntable method, and the error between these two methods is less than $10^{-3}$.…

3DMNDT 3D multi view registration method based on the normal distributions transform

The normal distributions transform (NDT) is an effective paradigm for the point set registration . This method is originally designed for pair-wiseregistration and it will suffer from great challenges when applied tomulti-view registration . The proposed method alternately implements data point clustering, NDT computing, and likelihood maximization until desired registration results are obtained .…