Back to results
US 2026/0278034 A1Published application
Systems And Methods For Predicting Future Events For A Temporal Point Process
AssigneeRoyal Bank of Canada
InventorsAmartya Mukherjee, Ruizhi Deng, He Zhao, Yuzhen Mao, Leonid Sigal, Frederick Tung
Life of this patent
Estimating the term…
Abstract
A diffusion model is used to generate events from a temporal point process (TPP) sequence. The diffusion model uses an asynchronous noise schedule. The TPP sequence is first mapped to a latent space and the diffusion model generates events in the latent space which are then mapped back to events.
Claims20
A method of generating a plurality of predicted events in a time series, the method comprising:
- receiving an observation series of events, z, comprising a plurality of events each associated with an event type of a plurality of defined event types and an event time, where:
- z
- =
- {
- z
- (
- 1
- )
- ,
- z
- (
- 2
- )
- ,
- …
- ,
- z
- (
- N
- )
- ;
- z
- (
- i
- )
- =
- (
- τ
- (
- i
- )
- ,
- k
- (
- i
- )
- )
- ;
- k(i) is the event type of the ith event;
- and
- τ(i) is an event time of the ith event;
- generating a latent event sequence, x, by encoding each event of the observation series of events to corresponding observation latent events in a latent space using a trained encoder, EØ, of a variational autoencoder, where:
- x
- =
- {
- x
- (
- 1
- )
- ,
- x
- (
- 2
- )
- ,
- …
- ,
- x
- (
- N
- )
- }
- ;
- x
- (
- 1
- )
- =
- E
- ∅
- (
- z
- (
- i
- )
- )
- generating a noise sample ϵ={ϵ1, . . . , ϵN+h}, where h is a number of events to predict;
- denoising the noise samples to generate a series of h latent events using a diffusion model conditioned on the received observation series x, the diffusion model applying an asynchronous noise schedule to denoise different samples at different rates;
- and
- decoding the h latent events to a prediction series of events using a trained decoder, Dψ, of the variational autoencoder.
77 more elements
A system comprising:
- a processor for executing instructions;
- and
- a memory storing instructions, which when executed by the processor configure the system to perform a method comprising:
- receiving an observation series of events, z, comprising a plurality of events each associated with an event type of a plurality of defined event types and an event time, where:
- z
- =
- {
- z
- (
- 1
- )
- ,
- z
- (
- 2
- )
- ,
- …
- ,
- z
- (
- N
- )
- ;
- z
- (
- i
- )
- =
- (
- τ
- (
- i
- )
- ,
- k
- (
- i
- )
- )
- ;
- k(i) is the event type of the ith event;
- and
- τ(i) is an event time of the ith event;
- generating a latent event sequence, x, by encoding each event of the observation series of events to corresponding observation latent events in a latent space using a trained encoder, EØ, of a variational autoencoder, where:
- x
- =
- {
- x
- (
- 1
- )
- ,
- x
- (
- 2
- )
- ,
- …
- ,
- x
- (
- N
- )
- }
- ;
- x
- (
- 1
- )
- =
- E
- ∅
- (
- z
- (
- i
- )
- )
- generating a noise sample ϵ={ϵ1, . . . , ϵN+h}, where h is a number of events to predict;
- denoising the noise samples to generate a series of h latent events using a diffusion model conditioned on the received observation series x, the diffusion model applying an asynchronous noise schedule to denoise different samples at different rates;
- and
- decoding the h latent events to a prediction series of events using a trained decoder, Dψ, of the variational autoencoder.
80 more elements
Dependencies are inferred from the claim text.