Back to results
US 11,295,207 B2Granted
Quantum deep learning
AssigneeMicrosoft Technology Licensing, LLC
InventorsNathan Wiebe, Krysta Svore, Ashish Kapoor
Life of this patent
Estimating the term…
Abstract
Boltzmann machines are trained using an objective function that is evaluated by sampling quantum states that approximate a Gibbs state. Classical processing is used to produce the objective function, and the approximate Gibbs state is based on weights and biases that are refined using the sample results. In some examples, amplitude estimation is used. A combined classical/quantum computer produces suitable weights and biases for classification of shapes and other applications.
Claims18
A method of efficiently training a Boltzmann machine, comprising:
- with a classical computer, receiving a specification of a Boltzmann machine, an objective function, and associated training data, and
- in a quantum computer, determining at least one gradient of the objective function by:
- based on the specification of the Boltzmann machine, preparing a plurality of qubits to represent a Gibbs distribution;
- adding a qubit to the plurality of qubits and applying a rotation operator so that the added qubit has amplitudes of first and second states based on a joint probability distribution of hidden and visible unit values;
- producing the at least one gradient of the objective function by sampling the states of each of the plurality of qubits;
- and
- based on the at least one gradient of the objective function, specifying at least one visible bias, at least one hidden bias or at least one weight of the Boltzmann machine so as to produce a trained Boltzmann machine.
4 more elements
A method, comprising:
- preparing at least one quantum state in a quantum computer to approximate a Gibbs state and sampling the at least one quantum state;
- preparing a qubit string storing energy values associated with a plurality of hidden weights and biases;
- adding a qubit to the prepared qubit string and applying a rotation operator so that the added qubit has amplitudes of first and second states based on a joint probability distribution of hidden and visible unit values;
- applying amplitude estimation to the prepared qubit string with the added qubit and measuring a state of the added qubit;
- determining a distribution of hidden and visible unit values based on the measured state of the added qubit;
- and
- estimating in a classical computer, gradients of an objective function based on the determined distribution of hidden and visible unit values.
4 more elements
A method, comprising:
- in a classical computer,
- receiving a training data set, a specification of a deep Boltzmann machine, and an objective function;
- in a quantum computer, establishing a mean-field state on a plurality of qubits based on a mean-field approximation so as to approximate a Gibbs state;
- and
- adding a qubit to the plurality of qubits and applying a rotation operator so that the added qubit has amplitudes based on a joint probability distribution of hidden and visible unit values;
- and
- in the classical computer,
- estimating a gradient of the objective function based on sampling the mean-field state, and determining hidden biases and weights of the deep Boltzmann machine;
- and
- processing an input data example associated with an image, a shape, speech data, a text, an audio recording, a video or a quantum state output by a quantum device based on the Boltzmann machine.
7 more elements
Dependencies are inferred from the claim text.