Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network. The neural network has a plurality of differentiable weights and a plurality of non-differentiable weights. One of the methods includes determining trained values of the plurality of differentiable weights and the non-differentiable weights by repeatedly performing operations that include determining an update to the current values of the plurality of differentiable weights using a machine learning gradient-based training technique and determining, using an evolution strategies (ES) technique, an update to the current values of a plurality of distribution parameters.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for executing a signal generation neural network on parallel processing hardware. One of the methods includes receiving weight matrices of a layer of a signal generation neural network. Rows of a first matrix for the layer are interleaved by assigning groups of rows of the first matrix to respective thread blocks of a plurality of thread blocks. A first subset of rows of the one or more other weight matrices are assigned to a first subset of the plurality of thread blocks and a second subset of rows of the one or more other weight matrices are assigned to a second subset of the plurality of thread blocks. The first matrix operation is performed substantially in parallel by the plurality of thread blocks. The other matrix operations are performed substantially in parallel by the plurality of thread blocks.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing a multivalue reduction using a parallel processing device. One of the methods includes performing a parallel M-value reduction by parallel processing units of a parallel processing device. A plurality of initial reductions are performed in serial, each initial reduction operating on data in a different respective register space of at least M register spaces. Data is moved from the M register spaces so that all results from the plurality of initial reductions are in a same first register space. One or more subsequent reductions are performed in parallel to compute M final values, each subsequent reduction operating only on data in the first register space.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output examples using neural networks. One of the methods includes receiving a request to generate an output example of a particular type, accessing dependency data, and generating the output example by, at each of a plurality of generation time steps: identifying one or more current blocks for the generation time step, wherein each current block is a block for which the values of the bits in all of the other blocks identified in the dependency for the block have already been generated; and generating the values of the bits in the current blocks for the generation time step conditioned on, for each current block, the already generated values of the bits in the other blocks identified in the dependency for the current block.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing a multivalue reduction using a parallel processing device. One of the methods includes performing a parallel M-value reduction by parallel processing units of a parallel processing device. A plurality of initial reductions are performed in serial, each initial reduction operating on data in a different respective register space of at least M register spaces. Data is moved from the M register spaces so that all results from the plurality of initial reductions are in a same first register space. One or more subsequent reductions are performed in parallel to compute M final values, each subsequent reduction operating only on data in the first register space.

Abstract: Methods and systems for arranging and searching a database of media content recordings are provided. In one example, a method is provided that comprises receiving a sample of media content, and performing, by a computing device, a content recognition of the sample of media content using a data file including a concatenation of representations for each of a plurality of media content recordings. In other examples, another method is provided that comprises receiving media content recordings, determining a representation for each media content recording, concatenating by a computing device the representation for each media content recording as a data file, and storing by the computing device a mapping between an identifier for a respective media content recording and a global position in the data file that corresponds to the representation of the respective media content recording.

Abstract: The invention provides an adapter for a dental milling block holder with a dental milling block (52 and 54) or dental grinding block, which is attached to a dental milling block holder, in particular glued on, wherein the dental milling block holder is provided with a round shank which is mounted in the adapter (40). It is characterized in that the adapter (40) is provided with at least one internal thread (92) for the mounting of a fastening screw (31) which acts on the shank, to which fastening screw (31) inside the shank mount of the adapter (40) of the fastening screw (31) positioned opposite of it, at least one in particular elevated, in particular at least two elevated clamping surfaces (74 and 76) are formed against which the shank abuts.

Abstract: Methods and systems for arranging and searching a database of media content recordings are provided. In one example, a method is provided that comprises receiving a sample of media content, and performing, by a computing device, a content recognition of the sample of media content using a data file including a concatenation of representations for each of a plurality of media content recordings. In other examples, another method is provided that comprises receiving media content recordings, determining a representation for each media content recording, concatenating by a computing device the representation for each media content recording as a data file, and storing by the computing device a mapping between an identifier for a respective media content recording and a global position in the data file that corresponds to the representation of the respective media content recording.

Abstract: The invention provides an adapter for a dental milling block holder with a dental milling block (52 and 54) or dental grinding block, which is attached to a dental milling block holder, in particular glued on, wherein the dental milling block holder is provided with a round shank which is mounted in the adapter (40). It is characterised in that the adapter (40) is provided with at least one internal thread (92) for the mounting of a fastening screw (31) which acts on the shank, to which fastening screw (31) inside the shank mount of the adapter (40) of the fastening screw (31) positioned opposite of it, at least one in particular elevated, in particular at least two elevated clamping surfaces (74 and 76) are formed against which the shank abuts.

Abstract: A holder for affixing a mill blank for the manufacture of a dental article in a housing of a milling unit of a CAD/CAM system, wherein the holder includes a joining surface for joining to the mill blank, a securing shaft having at least one securing surface for securing to a housing of a milling unit of a CAD/CAM system, wherein the securing shaft comprises an essentially polygonal shape with at least two lateral surfaces, wherein at least two of the lateral surfaces are arranged adjacent to each other and extend at right angles to each other; and wherein the securing surface, in cross-sectional view, extends substantially perpendicular to a radius of the securing shaft extending essentially through the center of the securing surface and comprises a width that is less than the width of a lateral surface.

Abstract: A holder for affixing a mill blank for the manufacture of a dental article in a housing of a milling unit of a CAD/CAM system, wherein the holder includes a joining surface for joining to the mill blank, a securing shaft having at least one securing surface for securing to a housing of a milling unit of a CAD/CAM system, wherein the securing shaft comprises an essentially polygonal shape with at least two lateral surfaces, wherein at least two of the lateral surfaces are arranged adjacent to each other and extend at right angles to each other; and wherein the securing surface, in cross-sectional view, extends substantially perpendicular to a radius of the securing shaft extending essentially through the center of the securing surface and comprises a width that is less than the width of a lateral surface.