International Business Machines Corporation

Globalfoundries Inc.

Eth Zurich (eidgenoessische Technische Hochschule Zurich)

Other

Eth Zurich

Giovanni Cherubini

Jens Jelitto

Mark Alfred Lantz

Simeon Furrer

Evangelos Stavros Eleftheriou

Nhan Xuan Bui

Robert Allan Hutchins

Roy Daron Cideciyan

Kazuhiro Tsuruta

Abu Sebastian

Robert Allen Hutchins

Angeliki Pantazi

Manuel Le Gallo-Bourdeau

Johan B C Engelen

Vinodh Venkatesan

Abbas Rahimi

Kumudu Geethan Karunaratne

Charalampos Pozidis

Paul T Hurley

Sedat Oelcer

Ilias Iliadis

Luca Benini

Thomas Helmut Toifl

Matthieu M Simeoni

Theodor W Loeliger

Reed Alan Hancock

Ajay Dholakia

Marcel A Kossel

Gerd K Binnig

Hazar Yüksel

Peter Baechtold

Sanaz Kazemi

Hazar Yuksel

Glen Alan Jaquette

Thomas R Albrecht

Sri Muthuthamby Sri-Jayantha

Michel Despont

Gottfried Ungerboeck

Johannes Georg Bednorz

Katherine Tyldesley Blinick

Theodore Antonakopoulos

Timoleon Moraitis

Orhan Ocal

Robert G Biskeborn

Thomas Mittelholzer

Malcolm Scott Ware

Eiji Ogura

Hugo Eric Rothuizen

Peter Vettiger

Randy Clark Inch

Tomoko Taketomi

Pierre-Olivier Jubert

Laurent A Dellmann

Mark Allan Taylor

Melanie Jean Sandberg

Manuel Le Gallo

Haris Pozidis

Yusik Kim

Gottfried Ungerbock

Urs T Dürig

Peter Bächtold

Evangelos S Eleftherious

Johan Bc Engelen

Hazar Yueksel

Taras Lehinevych

Ana Stanojevic

Sedat Oelcedillaer

Robert Hutchins

Evangelos Eleftheriou

Mark A Lantz

Robert A Hutchins

A computer-implemented method for answering a cognitive query from sensor input signals may be provided. The method comprises feeding sensor input signals to an input layer of an artificial neural network comprising a plurality of hidden neuron layers and an output neural layer, determining hidden layer output signals from each of the plurality of hidden neuron layers and output signals from the output neural layer, and generating a set of pseudo-random bit sequences by applying a set of mapping functions using the output signals of the output layer and the hidden layer output signals of one of the hidden neuron layers as input data for one mapping function. Furthermore, the method comprises determining a hyper-vector using the set of pseudo-random bit sequences, and storing the hyper-vector in an associative memory, in which a distance between different hyper-vectors is determinable.

Answering cognitive queries from sensor input signals

The present disclosure relates to a method for classifying a query information element using the similarity between the query information element and a set of support information elements. A resulting set of similarity scores is transformed using a sharpening function such that the transformed scores are decreasing as negative similarity scores increase and the transformed scores are increasing as positive similarity scores increase. A class of the query information element is determined based on the transformed similarity scores.

Memory-augmented neural network system

The present disclosure relates to a method for representing an ordered group of symbols with a hypervector. The method comprises sequentially applying on at least part of the input hypervector associated with a current symbol a predefined number of circular shift operations associated with the current symbol, resulting in a shifted hypervector. A rotate operation may be applied on the shifted hypervector, resulting in an output hypervector. If the current symbol is not the last symbol of the ordered group of symbols the output hypervector may be provided as the input hypervector associated with a subsequent symbol of the current symbol; otherwise, the output hypervector of the last symbol of the ordered group of symbols may be provided as a hypervector that represents the ordered group of symbols.

Representation of an ordered group of symbols by hypervectors

Methods and apparatus are provided for memorizing data signals in a spiking neural network. For each data signal, such a method includes supplying metadata relating to the data signal to a machine learning model trained to generate an output signal, indicating a relevance class for a data signal, from input metadata for that data signal. The method includes iteratively supplying the data signal to a sub-assembly of neurons, interconnected via synaptic weights, of a spiking neural network and training the synaptic weights to memorize the data signal in the sub-assembly. The method further comprises assigning neurons of the network to the sub-assembly in dependence on the output signal of the model such that more relevant data signals are memorized by larger sub-assemblies. The data signal memorized by a sub-assembly can be subsequently recalled by activating neurons of that sub-assembly.

Learning and recall in spiking neural networks

A computer-implemented method for managing datasets of a storage system is provided, wherein the datasets have respective sets of metadata, the method including: successively feeding first sets of metadata to a spiking neural network (SNN), the first sets of metadata fed corresponding to datasets of the storage system that are labeled with respect to classes they belong to, so as to be associated with class labels, for the SNN to learn representations of said classes in terms of connection weights that weight the metadata fed; successively feeding second sets of metadata to the SNN, the second sets of metadata corresponding to unlabeled datasets of the storage system, for the SNN to infer class labels for the unlabeled datasets, based on the second sets of metadata fed and the representations learned; and managing datasets in the storage system, based on class labels of the datasets, these including the inferred class labels.

Managing datasets of a cognitive storage system with a spiking neural network

A system for hyper-dimensional computing for inference tasks may be provided. The device comprises an item memory for storing hyper-dimensional item vectors, a query transformation unit connected to the item memory, the query transformation unit being adapted for forming a hyper-dimensional query vector from a query input and hyper-dimensional base vectors stored in the item memory, and an associative memory adapted for storing a plurality of hyper-dimensional profile vectors and for determining a distance between the hyper-dimensional query vector and the plurality of hyper-dimensional profile vectors, wherein the item memory and the associative memory are adapted for in-memory computing using memristive devices.

Device for hyper-dimensional computing tasks

A computer-implemented method for classification of an input element to an output class in a spiking neural network may be provided. The method comprises receiving an input data set comprising a plurality of elements, identifying a set of features and corresponding feature values for each element of the input data set, and associating each feature to a subset of spiking neurons of a set of input spiking neurons of the spiking neural network. Furthermore, the method comprises also generating, by the input spiking neurons, spikes at pseudo-random time instants depending on a value of the feature for a given input element, and classifying an element into a class depending on a distance measure value between output spiking patterns at output spiking neurons of the spiking neural network and a predefined target pattern related to the class.

Correlative time coding method for spiking neural networks

The invention is directed to a device for high-dimensional encoding of a plurality of sequences of quantitative data signals. The device comprises a plurality of input channels for receiving the plurality of sequences of quantitative data signals and an encoding unit. The encoding unit is configured to perform a temporal high-dimensional encoding of n-grams of the plurality of sequences of quantitative data signals; thereby creating a plurality of temporally encoded hypervectors for the plurality of input channels. The encoding unit is further configured to perform a spatial high-dimensional encoding of the plurality of temporally encoded hypervectors, thereby creating a temporally and spatially encoded hypervector. The device further comprises a configuration controller. The configuration controller is adapted to configure the high-dimensional encoding in dependence on one or more hyperparameter values.

Device for high dimensional encoding

The invention is notably directed at a device for high-dimensional computing comprising an associative memory module. The associative memory module comprises one or more planar crossbar arrays. The one or more planar crossbar arrays comprise a plurality of resistive memory elements. The device is configured to program profile vector elements of profile hypervectors as conductance states of the resistive memory elements and to apply query vector elements of query hypervectors as read voltages to the one or more crossbar arrays. The device is further configured to perform a distance computation between the profile hypervectors and the query hypervectors by measuring output current signals of the one or more crossbar arrays. The invention further concerns a related method and a related computer program product.

Device for high dimensional computing comprising an associative memory module

The invention is directed a device for high-dimensional encoding of a plurality of sequences of quantitative data signals. The device comprises a memory crossbar array comprising a plurality of resistive devices, a first peripheral circuit connected to the memory crossbar array, and a second peripheral circuit connected to the first peripheral circuit. The device is configured to receive the plurality of sequences of quantitative data signals via a plurality of input channels and to store elements of a plurality of precomputed basis hypervectors as conductance states of the resistive devices. The plurality of basis hypervectors are bound to respective input channels. The first peripheral circuit performs a temporal encoding of n-grams of the quantitative data signals thereby creating a plurality of temporally encoded hypervectors. The second peripheral circuit performs a spatial encoding of the plurality of temporally encoded hypervectors. This creates a temporally and spatially encoded hypervector.

Methods and apparatus are provided for calculating branch metrics, associated with possible transitions between states of a trellis, in a sequence detector for detecting symbol values corresponding to samples of an analog signal transmitted over a channel. For each sample and each transition, the method calculates a plurality of distance values indicative of distance between that sample and respective hypothesized sample values for that transition. In parallel with calculation of the distance values, the sample is compared with a set of thresholds, each defined between a pair of successive hypothesized symbol values arranged in value order, to produce a comparison result. An optimum distance value is selected as a branch metric for the transition in dependence on the comparison result.

Sequence detection

Methods and apparatus are provided for implementing propagation of probability distributions of random variables over a factor graph. Such a method includes providing a spiking neural network, having variable nodes interconnected with factor nodes, corresponding to the factor graph. Each of the nodes comprises a set of neurons configured to implement computational functionality of that node. The method further comprises generating, for each of a set of the random variables, at least one spike signal in which the probability of a possible value of that variable is encoded via the occurrence of spikes in the spike signal, and supplying the spike signals for the set of random variables as inputs to the neural network at respective variable nodes. The probability distributions are propagated via the occurrence of spikes in signals propagated through the neural network.

Probability propagation over factor graphs

A storage system and a method for storing a data segment, a storage capacity manager and a method for managing a capacity of a storage unit, and a storage tier relocation manager and a method for relocating a data segment. The storage system includes at least two storage tiers, an access pattern evaluator, a classification unit, a selector, and logic. The storage capacitor manager includes a monitoring unit and capacity managing unit. The storage tier relocation manager includes a target storage tier, the data segment relocated to the target storage tier, and a protection measure.

Automatically organizing storage system

A computer-implemented method, according to one embodiment, includes: receiving a first timestamp in response to a first servo reader detecting a stripe of a first servo burst on the magnetic tape, and receiving a second timestamp in response to a second servo reader detecting a stripe of a second servo burst on the magnetic tape. A third timestamp is received in response to the first servo reader detecting a stripe of a third servo burst on the magnetic tape, while a fourth timestamp is received in response to the second servo reader detecting a stripe of a fourth servo burst on the magnetic tape. The first, second, third, and fourth timestamps are used to determine the skew of the magnetic tape. Further still, the first and third servo bursts are in a same first servo sub-frame, while the second and fourth servo bursts are in a same second servo sub-frame.

Accurate skew determination for magnetic tapes experiencing the effects of tape dimensional instability

The present invention is notably directed to a computer-implemented method of accessing data by an application. The method comprises detecting an application that is accessing an initial file stored in a storage system; fetching an application model of the detected application, the application model comprising an association between access patterns of the detected application when accessing first files with first metadata information of the said first files; identifying one or more second files stored in the storage system, the identified one or more second files having second metadata information that are similar to the first metadata information; and prefetching the identified one or more second files.

Data prefetching for large data systems

Growing community of inventors

Rueschlikon, Switzerland

Giovanni Cherubini