Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a target action selection policy to control a target agent interacting with an environment. In one aspect, a method comprises: obtaining a set of offline training data, wherein the offline training data characterizes interaction of a baseline agent with an environment as the baseline agent performs actions selected in accordance with a baseline action selection policy; generating a set of online training data that characterizes interaction of the target agent with the environment as the target agent performs actions selected in accordance with the target action selection policy; and training the target action selection policy on both: (i) the offline training data, and (ii) the online training data, wherein the training of the target action selection policy on the offline training data is conditioned on a measure of competency of the baseline agent.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling an agent. One of the methods includes sampling one or more index variables from a continuous space of possible index variables in accordance with a probability distribution over the continuous space; for each index variable: processing the index variable using a hypermodel, in accordance with values of a plurality of parameters of the hypermodel, to generate an output that specifies values of a plurality of parameters of an environment model; and generating an action selection output using the environment model in accordance with the values of the plurality of parameters of the environment model that are specified by the hypermodel output for the index variable; and selecting the action to be performed by the agent at the time step using the one or more action selection outputs for the one or more index variables.

Abstract: Systems and methods for providing reinforcement learning for a deep learning network are disclosed. A reinforcement learning process that provides deep exploration is provided by a bootstrap that applied to a sample of observed and artificial data to facilitate deep exploration via a Thompson sampling approach.

Abstract: Systems and methods for providing reinforcement learning for a deep learning network are disclosed. A reinforcement learning process that provides deep exploration is provided by a bootstrap that applied to a sample of observed and artificial data to facilitate deep exploration via a Thompson sampling approach.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: In a mobile communication device, a method for compensating for a frequency adjustment in an oscillator shared between a communication circuit and a positioning signal receiver is provided. In one embodiment, the method begins to receive and store a positioning signal at a first time point. When, at a second time point, the operating frequency of the shared oscillator is adjusted, the frequency adjustment is recorded. After the positioning signal is completely received and stored, the processing of the positioning signal takes into consideration the frequency adjustment. In that embodiment, the processing hypothesizes a frequency shift in the received positioning signal. According to another aspect of the present invention, a method for determining the operating frequency of an oscillator detects a beginning time point of a reference signal received by the mobile communication device and enables a counter to count in step with a clock signal derived from the oscillator.

Abstract: In a mobile communication device, a method for compensating for a frequency adjustment in an oscillator shared between a communication circuit and a positioning signal receiver is provided. In one embodiment, a method for determining the operating frequency of an oscillator detects a beginning time point of a reference signal received by the mobile communication device and enables a counter to count in step with a clock signal derived from the oscillator. When an ending time point of the reference signal is received by the mobile communication device, the count is stopped, and the frequency of the oscillator is determined based on the count in the counter and an expected time that elapsed between the beginning time point and the ending time point.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: System and method to determine the location of a receiver in a multipath environment are provided. The received signal is correlated with the reference signals associated with the transmitting sources. Each correlation function is processed to derive various types of signal constraints, such as probability densities and uncertainty regions or intervals. In some embodiments, these constraints are for the code-phases and the Doppler frequencies. These signal constraints are transformed into constraints on the receiver variables and then fused together into a unified receiver constraint. A-priori constraints, such as constraints on the location of the receiver or the timestamp, may be incorporated into the unified receiver constraint. Some embodiments estimate a location based also on the estimated Doppler frequency. The constraints used by the invention are based on models of multipath effects and are geared towards mitigating these effects.

Abstract: Techniques are provided for fine-tuning estimates of a delay value for a sampled signal. One aspect of the invention is to perform, for the sampled signal, coarse-grained calculations of the In Phase and Quadrature (I and Q) correlation integrals at a limited number of points, wherein the calculations are performed over a range of hypothesized delay values. A range of delay values of interest are then determined from the coarse-grained calculations of the I and Q correlation integrals. A subset of I and Q values based on the coarse granularity calculations of the I and Q correlation functions is used to perform a time-domain interpolation to obtain fine-grained values of the I and Q integrals in the range of the delay values of interest. Magnitude calculations are performed based on the fine-grained values of the I and Q integrals. Fine-tuned estimates of delay value are based on the magnitude calculations. Alternatively, fine-tuned estimates of delay value are based on the template-matching approach.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: Techniques are provided for aiding in acquiring a signal using the data bit information that is associated with each signal source. One aspect of the invention is to use the data bit information that is associated with each signal source when calculating the In Phase and Quadrature correlation integrals by using the sampled data associated with the received signal. By using the data bit information that is associated with each signal source, coherent correlation may be performed by breaking the signal into data blocks and performing calculations on a block-by-block basis. Coherent correlation is the calculation of In Phase and Quadrature correlation integrals for sampled data that is associated with the received signal.

Abstract: In a mobile communication device, a method for compensating for a frequency adjustment in an oscillator shared between a communication circuit and a positioning signal receiver is provided. In one embodiment, the method begins to receive and store a positioning signal at a first time point. When, at a second time point, the operating frequency of the shared oscillator is adjusted, the frequency adjustment is recorded. After the positioning signal is completely received and stored, the processing of the positioning signal takes into consideration the frequency adjustment. In that embodiment, the processing hypothesizes a frequency shift in the received positioning signal. According to another aspect of the present invention, a method for determining the operating frequency of an oscillator detects a beginning time point of a reference signal received by the mobile communication device and enables a counter to count in step with a clock signal derived from the oscillator.

Abstract: A multi-function device with a positioning function and a real time positioning engine is disclosed. The device contains also a shared processor used by the positioning function and other functions of the device, such as a mobile radio-communication function. The positioning engine performs in real time the most computational intensive calculations of the positioning function, such as downsampling, Doppler mixing and correlation calculations. Since the received signal need not be stored, the memory requirements of the positioning function are significantly reduced if aiding data is available.

Abstract: In a mobile communication device, a method for compensating for a frequency adjustment in an oscillator shared between a communication circuit and a positioning signal receiver is provided. In one embodiment, the method begins to receive and store a positioning signal at a first time point. When, at a second time point, the operating frequency of the shared oscillator is adjusted, the frequency adjustment is recorded. After the positioning signal is completely received and stored, the processing of the positioning signal takes into consideration the frequency adjustment. In that embodiment, the processing hypothesizes a frequency shift in the received positioning signal. According to another aspect of the present invention, a method for determining the operating frequency of an oscillator detects a beginning time point of a reference signal received by the mobile communication device and enables a counter to count in step with a clock signal derived from the oscillator.

Abstract: System and method to determine the location of a receiver in a multipath environment are provided. The received signal is correlated with the reference signals associated with the transmitting sources. Each correlation function is processed to derive various types of signal constraints, such as probability densities and uncertainty regions or intervals. In some embodiments, these constraints are for the code-phases and the Doppler frequencies. These signal constraints are transformed into constraints on the receiver variables and then fused together into a unified receiver constraint. A-priori constraints, such as constraints on the location of the receiver or the timestamp, may be incorporated into the unified receiver constraint. Some embodiments estimate a location based also on the estimated Doppler frequency. The constraints used by the invention are based on models of multipath effects and are geared towards mitigating these effects.