Patents by Inventor Andreas Knoblauch

Andreas Knoblauch has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 8521669
    Abstract: This invention is in the field of machine learning and neural associative memory. In particular the invention discloses a neural associative memory structure for storing and maintaining associations between memory address patterns and memory content patterns using a neural network, as well as methods for storing and retrieving such associations. Bayesian learning is applied to achieve non-linear learning.
    Type: Grant
    Filed: May 28, 2010
    Date of Patent: August 27, 2013
    Assignee: Honda Research Institute Europe GmbH
    Inventor: Andreas Knoblauch
  • Patent number: 8463722
    Abstract: This invention is in the field of machine learning and neural associative memory. In particular the invention discloses a neural associative memory structure for storing and maintaining associations between memory address patterns and memory content patterns using a neural network, as well as methods for retrieving such associations. A method for a non-linear synaptic learning of discrete synapses is disclosed, and its application on neural networks is laid out.
    Type: Grant
    Filed: April 22, 2010
    Date of Patent: June 11, 2013
    Assignee: Honda Research Institute Europe GmbH
    Inventor: Andreas Knoblauch
  • Patent number: 8335752
    Abstract: A method for forming an associative computer memory comprises the step of forming an inhibitory memory matrix A?=?(Ap?A). According to the Wilshaw model constructed from a given set of address patterns and content patterns' and random matrix structure.
    Type: Grant
    Filed: June 20, 2008
    Date of Patent: December 18, 2012
    Assignee: Honda Research Institute Europe GmbH
    Inventor: Andreas Knoblauch
  • Patent number: 7875985
    Abstract: A memory device comprising at least one memory stack of stacked memory dies which are staggered with respect to each other, each stacked memory die of said memory stack comprising along its edge die pads for bonding said stacked memory die to substrate pads of said memory device connectable to a control circuit, wherein each die pad of a stacked memory die which connects said memory die individually to said control circuit comprises an increased distance (di) in comparison to die pads of said stacked memory die which connect said stacked memory die in parallel with corresponding die pads of other stacked memory dies of said memory stack to said control circuit.
    Type: Grant
    Filed: December 22, 2006
    Date of Patent: January 25, 2011
    Assignee: Qimonda AG
    Inventors: Dietmar Hiller, Roberto Dossi, Andreas Knoblauch
  • Publication number: 20100312731
    Abstract: This invention is in the field of machine learning and neural associative memory. In particular the invention discloses a neural associative memory structure for storing and maintaining associations between memory address patterns and memory content patterns using a neural network, as well as methods for storing and retrieving such associations. Bayesian learning is applied to achieve non-linear learning.
    Type: Application
    Filed: May 28, 2010
    Publication date: December 9, 2010
    Applicant: HONDA RESEARCH INSTITUTE EUROPE GMBH
    Inventor: Andreas KNOBLAUCH
  • Publication number: 20100312735
    Abstract: This invention is in the field of machine learning and neural associative memory. In particular the invention discloses a neural associative memory structure for storing and maintaining associations between memory address patterns and memory content patterns using a neural network, as well as methods for retrieving such associations. A method for a non-linear synaptic learning of discrete synapses is disclosed, and its application on neural networks is laid out.
    Type: Application
    Filed: April 22, 2010
    Publication date: December 9, 2010
    Applicant: HONDA RESEARCH INSTITUTE EUROPE GMBH
    Inventor: Andreas Knoblauch
  • Publication number: 20100306145
    Abstract: A method for forming an associative computer memory comprises the step of forming an inhibitory memory matrix A?=?(Ap?A). According to the Wilshaw model constructed from a given set of address patterns and content patterns' and random matrix structure.
    Type: Application
    Filed: June 20, 2008
    Publication date: December 2, 2010
    Applicant: HONDA RESEARCH INSTITUTE EUROPE GMBH
    Inventor: Andreas Knoblauch
  • Publication number: 20080150111
    Abstract: A memory device comprising at least one memory stack of stacked memory dies which are staggered with respect to each other, each stacked memory die of said memory stack comprising along its edge die pads for bonding said stacked memory die to substrate pads of said memory device connectable to a control circuit, wherein each die pad of a stacked memory die which connects said memory die individually to said control circuit comprises an increased distance (di) in comparison to die pads of said stacked memory die which connect said stacked memory die in parallel with corresponding die pads of other stacked memory dies of said memory stack to said control circuit.
    Type: Application
    Filed: December 22, 2006
    Publication date: June 26, 2008
    Inventors: Dietmar Hiller, Roberto Dossi, Andreas Knoblauch