专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO2018211144A1 MAKING OBJECT-LEVEL PREDICTIONS OF THE FUTURE STATE OF A PHYSICAL SYSTEM 审中-公开
公开(公告)号：WO2018211144A1
公开(公告)日：2018-11-22
申请号：PCT/EP2018/063304
申请日：2018-05-22
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： WATTERS, Nicholas , PASCANU, Razvan , BATTAGLIA, Peter William , ZORN, Daniel , WEBER, Theophane Guillaume
IPC分类号： G06N3/04 , G06N3/08
摘要： A system implemented by one or more computers comprises a visual encoder component configured to receive as input data representing a sequence of image frames, in particular representing objects in a scene of the sequence, and to output a sequence of corresponding state codes, each state code comprising vectors, one for each of the objects. Each vector represents a respective position and velocity of its corresponding object. The system also comprises a dynamic predictor component configured to take as input a sequence of state codes, for example from the visual encoder, and predict a state code for a next unobserved frame. The system further comprises a state decoder component configured to convert the predicted state code, to a state, the state comprising a respective position and velocity vector for each object in the scene. This state may represent a predicted position and velocity vector for each of the objects.

2. 发明申请

WO2018071392A1 NEURAL NETWORKS FOR SELECTING ACTIONS TO BE PERFORMED BY A ROBOTIC AGENT 审中-公开
标题翻译：用机器人进行动作选择的神经网络
公开(公告)号：WO2018071392A1
公开(公告)日：2018-04-19
申请号：PCT/US2017/055894
申请日：2017-10-10
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： PASCANU, Razvan , HADSELL, Raia Thais , VECERIK, Matej , ROTHORL, Thomas , RUSU, Andrei-Alexandru , HEESS, Nicolas Manfred Otto
IPC分类号： G06N3/04 , G06N3/00 , G06N3/08
摘要： A system includes a neural network system implemented by one or more computers. The neural network system is configured to receive an observation characterizing a current state of a real-world environment being interacted with by a robotic agent to perform a robotic task and to process the observation to generate a policy output that defines an action to be performed by the robotic agent in response to the observation. The neural network system includes: (i) a sequence of deep neural networks (DNNs), in which the sequence of DNNs includes a simulation-trained DNN that has been trained on interactions of a simulated version of the robotic agent with a simulated version of the real-world environment to perform a simulated version of the robotic task, and (ii) a first robot-trained DNN that is configured to receive the observation and to process the observation to generate the policy output.
摘要翻译：系统包括由一个或多个计算机实现的神经网络系统。神经网络系统被配置为接收表征机器人代理正在与之交互的真实世界环境的当前状态的观察结果，以执行机器人任务并处理该观察结果以生成策略输出，该输出定义将由响应观察的机器人代理。该神经网络系统包括：（i）深度神经网络（DNN）序列，其中DNN序列包括经过模拟训练的DNN，该模拟训练的DNN已经在模拟版本的机器人代理与模拟版本的执行模拟版本的机器人任务的真实世界环境，以及（ii）第一机器人训练的DNN，其被配置为接收观察并处理观察以生成策略输出。

3. 发明申请

WO2019202073A1 NEURAL NETWORKS FOR SCALABLE CONTINUAL LEARNING IN DOMAINS WITH SEQUENTIALLY LEARNED TASKS 审中-公开
公开(公告)号：WO2019202073A1
公开(公告)日：2019-10-24
申请号：PCT/EP2019/060070
申请日：2019-04-18
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： SCHWARZ, Jonathan , PASCANU, Razvan , HADSELL, Raia Thais , CZARNECKI, Wojciech , THE, Yee Whye , LUKETINA, Jelena
IPC分类号： G06N3/04 , G06N3/08
摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for scalable continual learning using neural networks. One of the methods includes receiving new training data for a new machine learning task; training an active subnetwork on the new training data to determine trained values of the active network parameters from initial values of the active network parameters while holding current values of the knowledge parameters fixed; and training a knowledge subnetwork on the new training data to determine updated values of the knowledge parameters from the current values of the knowledge parameters by training the knowledge subnetwork to generate knowledge outputs for the new training inputs that match active outputs generated by the trained active subnetwork for the new training inputs.

4. 发明申请

WO2019081782A1 MACHINE LEARNING SYSTEMS WITH MEMORY BASED PARAMETER ADAPTATION FOR LEARNING FAST AND SLOWER 审中-公开
公开(公告)号：WO2019081782A1
公开(公告)日：2019-05-02
申请号：PCT/EP2018/079559
申请日：2018-10-29
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： SPRECHMANN, Pablo , JAYAKUMAR, Siddhant , RAE, Jack William , PRITZEL, Alexander , BADIA, Adrià Puigdomènech , VINYALS, Oriol , PASCANU, Razvan , BLUNDELL, Charles
IPC分类号： G06N3/04 , G06N3/08
CPC分类号： G06N3/0454 , G06N3/006 , G06N3/0445 , G06N3/084
摘要： There is described herein a computer-implemented method of processing an input data item. The method comprises processing the input data item using a parametric model to generate output data, wherein the parametric model comprises a first sub-model and a second sub-model. The processing comprises processing, by the first sub-model, the input data to generate a query data item, retrieving, from a memory storing data point-value pairs, at least one data point-value pair based upon the query data item and modifying weights of the second sub-model based upon the retrieved at least one data point-value pair. The output data is then generated based upon the modified second sub-model.

5. 发明申请

WO2018211138A8 MULTITASK NEURAL NETWORK SYSTEMS WITH TASK-SPECIFIC POLICIES AND A SHARED POLICY 审中-公开
公开(公告)号：WO2018211138A8
公开(公告)日：2018-11-22
申请号：PCT/EP2018/063275
申请日：2018-05-22
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： PASCANU, Razvan , HADSELL, Raia Thais , BAPST, Victor Constant , CZARNECKI, Wojciech , KIRKPATRICK, James , TEH, Yee Whye , HEESS, Nicolas Manfred Otto
IPC分类号： G06N3/04 , G06N3/00 , G06N3/08
摘要： A method is proposed for training a multitask computer system, such as a multitask neural network system. The system comprises a set of trainable workers and a shared module. The trainable workers and shared module are trained on a plurality of different tasks, such that each worker learns to perform a corresponding one of the tasks according to a respective task policy, and said shared policy network learns a multitask policy which represents common behavior for the tasks. The coordinated training is performed by optimizing an objective function comprising, for each task: a reward term indicative of an expected reward earned by a worker in performing the corresponding task according to the task policy; and at least one entropy term which regularizes the distribution of the task policy towards the distribution of the multitask policy.

6. 发明申请

WO2022069747A1 TRAINING REINFORCEMENT LEARNING AGENTS USING AUGMENTED TEMPORAL DIFFERENCE LEARNING 审中-公开
公开(公告)号：WO2022069747A1
公开(公告)日：2022-04-07
申请号：PCT/EP2021/077182
申请日：2021-10-01
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： GULCEHRE, Caglar , PASCANU, Razvan , GOMEZ, Sergio
IPC分类号： G06N3/00 , G06N3/04 , G06N3/08
摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network used to select actions performed by an agent interacting with an environment by performing actions that cause the environment to transition states. One of the methods includes maintaining a replay memory storing a plurality of transitions; selecting a plurality of transitions from the replay memory; and training the neural network on the plurality of transitions, comprising, for each transition: generating an initial Q value for the transition; determining a scaled Q value for the transition; determining a scaled temporal difference learning target for the transition; determining an error between the scaled temporal difference learning target and the scaled Q value; determining an update to the current values of the Q network parameters; and determining an update to the current value of the scaling term.

7. 发明申请

WO2021245275A1 SYSTEM AND METHOD FOR TRAINING A SPARSE NEURAL NETWORK WHILST MAINTAINING SPARSITY 审中-公开
公开(公告)号：WO2021245275A1
公开(公告)日：2021-12-09
申请号：PCT/EP2021/065069
申请日：2021-06-04
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： JAYAKUMAR, Siddhant Madhu , PASCANU, Razvan , RAE, Jack William , OSINDERO, Simon , ELSEN, Erich Konrad
IPC分类号： G06N3/08 , G06N3/04 , G06N3/0445 , G06N3/0454 , G06N3/082 , G06N3/084
摘要： A computer-implemented method of training a neural network. The method comprises repeatedly determining a forward-pass set of network parameters by selecting a first sub-set of parameters of the neural network and setting all other parameters of the forward-pass set of network parameters to zero. The method then processes a training data item using the neural network in accordance with the forward-pass set of network parameters to generate a neural network output, determines a value of an objective function from the neural network output and the training data item, selects a second sub-set of network parameters, determines a backward-pass set of network parameters comprising the first and second sub-sets of parameters, and updates parameters corresponding to the backward-pass set of network parameters using a gradient estimate determined from the value of the objective function.

8. 发明申请

WO2021058270A1 GATED ATTENTION NEURAL NETWORKS 审中-公开
公开(公告)号：WO2021058270A1
公开(公告)日：2021-04-01
申请号：PCT/EP2020/074913
申请日：2020-09-07
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： PARISOTTO, Emilio , SONG, Hasuk , RAE, Jack William , JAYAKUMAR, Siddhant Madhu , JADERBERG, Maxwell Elliot , PASCANU, Razvan , GULCEHRE, Caglar
IPC分类号： G06N3/04
摘要： A system including an attention neural network that is configured to receive an input sequence and to process the input sequence to generate an output is described. The attention neural network includes: an attention block configured to receive a query input, a key input, and a value input that are derived from an attention block input. The attention block includes an attention neural network layer configured to: receive an attention layer input derived from the query input, the key input, and the value input, and apply an attention mechanism to the query input, the key input, and the value input to generate an attention layer output for the attention neural network layer; and a gating neural network layer configured to apply a gating mechanism to the attention block input and the attention layer output of the attention neural network layer to generate a gated attention output.

9. 发明申请

WO2018211142A1 IMAGINATION-BASED AGENT NEURAL NETWORKS 审中-公开
公开(公告)号：WO2018211142A1
公开(公告)日：2018-11-22
申请号：PCT/EP2018/063291
申请日：2018-05-22
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： WIERSTRA, Daniel Pieter , LI, Yujia , PASCANU, Razvan , BATTAGLIA, Peter William , WEBER, Theophane Guillaume , BUESING, Lars , REICHERT, David Paul , VINYALS, Oriol , HEESS, Nicolas Manfred Otto , RACANIERE, Sebastien Henri
IPC分类号： G06N3/04 , G06N3/08 , G06N3/00
摘要： A neural network system is proposed to select actions to be performed by an agent interacting with an environment to perform a task in an attempt to achieve a specified result. The system may include a controller to receive state data and context data, and to output action data. The system may also include an imagination module to receive the state and action data, and to output consequent state data. The system may also include a manager to receive the state data and the context data, and to output route data which defines whether the system is to execute an action or to imagine. The system may also include a memory to store the context data.

10. 发明申请

WO2018211141A1 IMAGINATION-BASED AGENT NEURAL NETWORKS 审中-公开
公开(公告)号：WO2018211141A1
公开(公告)日：2018-11-22
申请号：PCT/EP2018/063283
申请日：2018-05-22
申请人： DEEPMIND TECHNOLOGIES LIMITED
发明人： WIERSTRA, Daniel Pieter , LI, Yujia , PASCANU, Razvan , BATTAGLIA, Peter William , WEBER, Theophane Guillaume , BUESING, Lars , REICHERT, David Paul , GUEZ, Arthur Clement , REZENDE, Danilo Jiminez , BADIA, Adria Puigdomenech , VINYALS, Oriol , HEESS, Nicolas Manfred Otto , RACANIERE, Sebastien Henri
IPC分类号： G06N3/04 , G06N3/08 , G06N3/00
摘要： A neural network system is proposed. The network can be trained by model-based reinforcement learning to select actions to be performed by an agent interacting with an environment, to perform a task in an attempt to achieve a specified result. The system may comprise at least one imagination core which has an input to receive a current observation characterizing a current state of the environment, and optionally historical observations, and which includes a model of the environment. The imagination core may be configured to output trajectory data in response to the current observation, and/or historical observations. The trajectory data comprising a sequence of future features of the environment imagined by the imagination core. The system may also include at least one rollout encoder to encode the sequence of features, and a reinforcement learning output stage to receive data derived from the rollout embedding and to output action policy data for defining an action policy identifying an action based on the current observation.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式