-
21.发明授权公开(公告)号：US11042811B2
公开(公告)日：2021-06-22
申请号：US15725600
申请日：2017-10-05
Applicant: D-Wave Systems Inc.
Inventor： Jason Rolfe , William G. Macready , Zhengbing Bian , Fabian A. Chudak
IPC: G06N10/00 , G06N3/04 , G06K9/00 , G06N3/08 , G06F15/80 , G06N20/00 , G06K9/62 , G06N20/10 , G06N7/00
Abstract: A computational system can include digital circuitry and analog circuitry, for instance a digital processor and a quantum processor. The quantum processor can operate as a sample generator providing samples. Samples can be employed by the digital processing in implementing various machine learning techniques. For example, the computational system can perform unsupervised learning over an input space, for example via a discrete variational auto-encoder, and attempting to maximize the log-likelihood of an observed dataset. Maximizing the log-likelihood of the observed dataset can include generating a hierarchical approximating posterior. Unsupervised learning can include generating samples of a prior distribution using the quantum processor. Generating samples using the quantum processor can include forming chains of qubits and representing discrete variables by chains.
-
公开(公告)号：US20210089884A1
公开(公告)日：2021-03-25
申请号：US16772094
申请日：2018-12-12
Applicant: D-WAVE SYSTEMS INC.
Inventor： William G. Macready , Jason T. Rolfe
Abstract: Collaborative filtering systems based on variational autoencoders (VAEs) are provided. VAEs may be trained on row-wise data without necessarily training a paired VAE on column-wise data (or vice-versa), and may optionally be trained via minibatches. The row-wise VAE models the output of the corresponding column-based VAE as a set of parameters and uses these parameters in decoding. In some implementations, a paired VAE is provided which receives column-wise data and models row-wise parameters; each of the paired VAEs may bind their learned column- or row-wise parameters to the output of the corresponding VAE. The paired VAEs may optionally be trained via minibatches. Unobserved data may be explicitly modelled. Methods for performing inference with such VAE-based collaborative filtering systems are also disclosed, as are example applications to search and anomaly detection.
-
公开(公告)号：US20200210876A1
公开(公告)日：2020-07-02
申请号：US15753661
申请日：2016-08-18
Applicant: D-Wave Systems Inc.
Inventor： Jason Rolfe , Dmytro Korenkevych , Mani Ranjbar , Jack R. Raymond , William G. Macready
Abstract: A computational system can include digital circuitry and analog circuitry, for instance a digital processor and a quantum processor. The quantum processor can operate as a sample generator providing samples. Samples can be employed by the digital processing in implementing various machine learning techniques. For example, the digital processor can operate as a restricted Boltzmann machine. The computational system can operate as a quantum-based deep belief network operating on a training data-set.
-
公开(公告)号：US20200160175A1
公开(公告)日：2020-05-21
申请号：US16682976
申请日：2019-11-13
Applicant: D-WAVE SYSTEMS INC.
Inventor： Arash Vahdat , Mostafa S. Ibrahim , William G. Macready
Abstract: Fully-supervised semantic segmentation machine learning models are augmented by ancillary machine learning models which generate high-detail predictions from low-detail, weakly-supervised data. The combined model can be trained over both fully- and weakly-supervised data. Only the primary model is required for inference, post-training. The combined model can be made self-correcting during training by adjusting the ancillary model's output based on parameters learned over both the fully- and weakly-supervised data. The self-correction module may combine the output of the primary and ancillary models in various ways, including through linear combinations and via neural networks. The self-correction module and ancillary model may benefit from disclosed pre-training techniques.
-
25.发明申请公开(公告)号：US20200090050A1
公开(公告)日：2020-03-19
申请号：US16562192
申请日：2019-09-05
Applicant: D-WAVE SYSTEMS INC.
Inventor： Jason T. Rolfe , Seyed Ali Saberali , William G. Macready
Abstract: Generative machine learning models, such as variational autoencoders, with comparatively sparse latent spaces are provided. Continuous latent variables are activated and/or inactivated based on a state of the latent space. Activation may be controlled by corresponding binary latent variables and/or by rectification of probability distributions defined over the latent space. Sparsification may be supported by normalization of terms, such as providing an L1 or L2 prior.
-
Patent Agency Ranking
26.发明申请公开(公告)号：US20170178017A1
公开(公告)日：2017-06-22
申请号：US15338915
申请日：2016-10-31
Applicant: D-Wave Systems Inc.
Inventor： Aidan Patrick Roy , William G. Macready
Abstract: Systems and methods allow formulation of embeddings of problems via targeted hardware (e.g., particular quantum processor). In a first stage, sets of connected subgraphs are successively generated, each set including a respective subgraph for each decision variable in the problem graph, adjacent decisions variables in the problem graph mapped to respective vertices in the hardware graph, the respective vertices which are connected by at least one respective edge in the hardware graph. In a second stage, the connected subgraphs are refined such that no vertex represents more than a single decision variable.
-
公开(公告)号：US20170177751A1
公开(公告)日：2017-06-22
申请号：US15419083
申请日：2017-01-30
Applicant: D-Wave Systems Inc.
Inventor： William G. Macready , Geordie Rose , Thomas F.W. Mahon , Peter Love , Marshall Drew-Brook
CPC classification number: G06F17/505 , B82Y10/00 , G06F17/11 , G06N99/002
Abstract: Solving computational problems may include generating a logic circuit representation of the computational problem, encoding the logic circuit representation as a discrete optimization problem, and solving the discrete optimization problem using a quantum processor. Output(s) of the logic circuit representation may be clamped such that the solving involves effectively executing the logic circuit representation in reverse to determine input(s) that corresponds to the clamped output(s). The representation may be of a Boolean logic circuit. The discrete optimization problem may be composed of a set of miniature optimization problems, where each miniature optimization problem encodes a respective logic gate from the logic circuit representation. A quantum processor may include multiple sets of qubits, each set coupled to respective annealing signal lines such that dynamic evolution of each set of qubits is controlled independently from the dynamic evolutions of the other sets of qubits.
-
公开(公告)号：US09396440B2
公开(公告)日：2016-07-19
申请号：US13796949
申请日：2013-03-12
Applicant: D-Wave Systems Inc.
Inventor： William G. Macready , Edward D. Dahl
CPC classification number: G06N99/002 , B82Y10/00 , G06N5/003 , G06N5/04
Abstract: Systems and methods to solve combinatorial problems employ a permutation network which may be modeled after a sorting network where comparators are replaced by switches that controllably determine whether inputs are swapped or are left unchanged at the outputs. A quantum processor may be used to generate permutations by the permutation network by mapping the state of each switch in the network to the state of a respective qubit in the quantum processor. In this way, a quantum computation may explore all possible permutations simultaneously to identify a permutation that satisfies at least one solution criterion. The Travelling Salesman Problem is discussed as an example of a combinatorial problem that may be solved using these systems and methods.
Abstract translation: 用于解决组合问题的系统和方法采用置换网络，其可以在分类网络之后建模，其中比较器被可控地确定输入是否被交换或在输出处保持不变的开关替代。 量子处理器可以用于通过将网络中每个交换机的状态映射到量子处理器中相应量子位的状态来通过置换网络来产生置换。 以这种方式，量子计算可以同时探索所有可能的排列以识别满足至少一个解决标准的置换。 讨论旅行销售员问题作为可以使用这些系统和方法解决的组合问题的示例。
-
公开(公告)号：US20130282636A1
公开(公告)日：2013-10-24
申请号：US13796949
申请日：2013-03-12
Applicant: D-WAVE SYSTEMS INC.
Inventor： William G. Macready , Edward D. Dahl
IPC: G06N99/00
CPC classification number: G06N99/002 , B82Y10/00 , G06N5/003 , G06N5/04
Abstract: Systems and methods to solve combinatorial problems employ a permutation network which may be modeled after a sorting network where comparators are replaced by switches that controllably determine whether inputs are swapped or are left unchanged at the outputs. A quantum processor may be used to generate permutations by the permutation network by mapping the state of each switch in the network to the state of a respective qubit in the quantum processor. In this way, a quantum computation may explore all possible permutations simultaneously to identify a permutation that satisfies at least one solution criterion. The Travelling Salesman Problem is discussed as an example of a combinatorial problem that may be solved using these systems and methods.
Abstract translation: 用于解决组合问题的系统和方法采用置换网络，其可以在分类网络之后建模，其中比较器被可控地确定输入是否被交换或在输出处保持不变的开关替代。 量子处理器可以用于通过将网络中每个交换机的状态映射到量子处理器中相应量子位的状态来通过置换网络来产生置换。 以这种方式，量子计算可以同时探索所有可能的排列以识别满足至少一个解决标准的置换。 讨论旅行销售员问题作为可以使用这些系统和方法解决的组合问题的示例。
-
30.发明授权公开(公告)号：US11704586B2
公开(公告)日：2023-07-18
申请号：US17739411
申请日：2022-05-09
Applicant: D-WAVE SYSTEMS INC.
Inventor： Murray C. Thom , Aidan P. Roy , Fabian A. Chudak , Zhengbing Bian , William G. Macready , Robert B. Israel , Kelly T. R. Boothby , Sheir Yarkoni , Yanbo Xue , Dmytro Korenkevych
IPC: G06N10/00
CPC classification number: G06N10/00
Abstract: Computational systems implement problem solving using hybrid digital/quantum computing approaches. A problem may be represented as a problem graph which is larger and/or has higher connectivity than a working and/or hardware graph of a quantum processor. A quantum processor may be used determine approximate solutions, which solutions are provided as initial states to one or more digital processors which may implement classical post-processing to generate improved solutions. Techniques for solving problems on extended, more-connected, and/or “virtual full yield” variations of the processor's actual working and/or hardware graphs are provided. A method of operation in a computational system comprising a quantum processor includes partitioning a problem graph into sub-problem graphs, and embedding a sub-problem graph onto the working graph of the quantum processor. The quantum processor and a non-quantum processor-based device generate partial samples. A controller causes a processing operation on the partial samples to generate complete samples.
-
-
-
-
-
-
-
-
-
Search Results
52
records
(took 0.025 seconds)
