Search

We consider the setting of a Master server, M, who possesses confidentialdata (e.g., personal, genomic or medical data) and wants to run intensivecomputations on it, as part of a machine learning algorithm for example. TheMaster wants to distribute these computations to untrusted workers who...

The paper deals with learning the probability distribution of the observeddata by artificial neural networks. We suggest a so-called gradient conjugateprior (GCP) update appropriate for neural networks, which is a modification ofthe classical Bayesian update for conjugate priors. We establish a...

All current popular hand-crafted key-point detectors such as Harris corner,MSER, SIFT, SURF... rely on some specific pre-designed structures for thedetection of corners, blobs, or junctions in an image. In this paper, a novelsparse coding based key point detector which requires no particularpre...

Active Internet measurement studies rely on a list of targets to be scanned.While probing the entire IPv4 address space is feasible for scans of limitedcomplexity, more complex scans do not scale to measuring the full Internet.Thus, a sample of the Internet can be used instead, often in form of a...

In this work, we present an alternative distribution layer for Erlang, namedPartisan. Partisan is a topology-agnostic distributed programming model anddistribution layer that supports several network topologies for differentapplication scenarios: full mesh, peer-to-peer, client-server, andpublish...

The performance of automatic speech recognition systems degrades withincreasing mismatch between the training and testing scenarios. Differences inspeaker accents are a significant source of such mismatch. The traditionalapproach to deal with multiple accents involves pooling data from...

Instructional design is a fundamental base for educational technologies as itlays the foundation to facilitate learning and teaching based on pedagogicalunderpinnings. However, most of the educational technologies today face twocore challenges in this context: (i) lack of instructional design as a...

One of the biggest challenges in the research of generative adversarialnetworks (GANs) is assessing the quality of generated samples and detectingvarious levels of mode collapse. In this work, we construct a novel measure ofperformance of a GAN by comparing geometrical properties of the underlying...

A divide and conquer strategy for enhancement of noisy speeches in adverseenvironments involving lower levels of SNR is presented in this paper, wherethe total system of speech enhancement is divided into two separate steps. Thefirst step is based on noise compensation on short time magnitude and...

There is a vast body of work on the capacity bounds for a "coherent" wirelessnetwork, where the network channel gains are known, at least at thedestination. However, there has been much less attention to the case where thenetwork parameters (channel gains) are unknown to everyone, i.e., thenon...

We perform fine-grained land use mapping at the city scale using ground-levelimages. Mapping land use is considerably more difficult than mapping land coverand is generally not possible using overhead imagery as it requires close-upviews and seeing inside buildings. We postulate that the growing...

We present PPFNet - Point Pair Feature NETwork for deeply learning a globallyinformed 3D local feature descriptor to find correspondences in unorganizedpoint clouds. PPFNet learns local descriptors on pure geometry and is highlyaware of the global context, an important cue in deep learning. Our...

Exploiting the efficiency and stability of Position-Based Dynamics (PBD), weintroduce a novel crowd simulation method that runs at interactive rates forhundreds of thousands of agents. Our method enables the detailed modeling ofper-agent behavior in a Lagrangian formulation. We model short-range...

Fish in schooling formations navigate complex flow-fields replete withmechanical energy in the vortex wakes of their companions. Their schoolingbehaviour has been associated with evolutionary advantages including collectiveenergy savings. How fish harvest energy from their complex fluid...

The insect olfactory system, which includes the antennal lobe (AL), mushroombody (MB), and ancillary structures, is a relatively simple neural systemcapable of learning. Its structural features, which are widespread inbiological neural systems, process olfactory stimuli through a cascade ofnetworks...

The recent success of deep neural networks is powered in part by large-scalewell-labeled training data. However, it is a daunting task to laboriouslyannotate an ImageNet-like dateset. On the contrary, it is fairly convenient,fast, and cheap to collect training images from the Web along with their...

It is undeniable that most developers today are building distributedapplications. However, most of these applications are developed by composingexisting systems together through unspecified APIs exposed to the applicationdeveloper. Systems are not going away: they solve a particular problem and...

A Dirichlet $k$-partition of a closed $d$-dimensional surface is a collectionof $k$ pairwise disjoint open subsets such that the sum of their firstLaplace-Beltrami-Dirichlet eigenvalues is minimal. In this paper, we develop asimple and efficient diffusion generated method to compute Dirichlet$k$...

The exponential explosion of parallel interleavings remains a fundamentalchallenge to model checking of concurrent programs. Both partial-orderreduction (POR) and transaction reduction (TR) decrease the number ofinterleavings in a concurrent system. Unlike POR, transactions also reduce thenumber of...

The recently introduced Thermodynamic Binding Networks (TBN) model wasdeveloped with the purpose of studying self-assembling systems by focusing ontheir thermodynamically favorable final states, and ignoring the kineticpathways through which they evolve. The model was intentionally developed...