Controllable dynamic multi-task architectures
AssigneeNEC Corporation
InventorsYumin Suh, Samuel Schulter, Xiang Yu, Masoud Faraki, Manmohan Chandraker, Dripta Raychaudhuri
Life of this patent
Estimating the term…
Abstract
Systems and methods for generating a hypernetwork configured to be trained for a plurality of tasks; receiving a task preference vector identifying a hierarchical priority for the plurality of tasks, and a resource constraint as a tuple; finding tree sub-structures and the corresponding modulation of features for every tuple within an N-stream anchor network; optimizing a branching regularized loss function to train an edge hypernet; and training a weight hypernet, keeping the anchor net and the edge hypernet fixed.
Claims20
A method for building a dynamic multi-task network, comprising:
- generating a hypernetwork configured to model tradeoffs among a plurality of competing tasks;
- receiving a task preference vector identifying a hierarchical priority for the plurality of competing tasks, and a resource constraint as a tuple;
- finding tree sub-structures and the corresponding modulation of features for every tuple within an N-stream anchor network;
- optimizing a branching regularized loss function to train an edge hypernet, wherein the branching regularized loss function is based on an active loss Lactive and an inactive loss, the active loss, Lactive, is calculated as:
- ℒ
- active
- =
- ∑
- l
- =
- 1
- L
-
- ∑
- i
- ,
- j
- ∈
- 𝒜
- i
- ≠
- j
-
- L
- -
- l
- L
- ·
- A
-
- (
- i
- ,
- j
- )
- ·
- P
-
- (
- l
- ,
- i
- ,
- j
- )
- ·
-
- v
- i
- l
- -
- v
- j
- l
-
- 2
- where P(l, i, j)=Puse(l, i)>Puse(l, j) is the probability that the nodes i and j in layer l are used in the sampled tree structure, and A(i, j) captures the task affinity between tasks, τi and τj, and
- υ
- i
- l
- and
- υ
- j
- l
- are samples, represents active tasks and L represents the total number of layers;
- training a weight hypernet, keeping the anchor network and the edge hypernet fixed;
- and
- generating, by the hypernetwork before inference time, an architectural configuration for a main model based on the task preference vector and the resource constraint.
A computer system for building a dynamic multi-task network, comprising:
- one or more processors;
- a memory in communication with the one or more processors;
- and
- a dynamic multi-task network builder stored in the memory, wherein the dynamic multi-task network builder performs;
- generate a hypernetwork configured to model tradeoffs among a plurality of competing tasks;
- receive a task preference vector identifying a hierarchical priority for the plurality of competing tasks, and a resource constraint as a tuple;
- find tree sub-structures and the corresponding modulation of features for every tuple within an N-stream anchor network;
- optimize a branching regularized loss function to train an edge hypernet, wherein the branching regularized loss function is based on an active loss Lactive and an inactive loss, the active loss, Lactive, is calculated as:
- ℒ
- active
- =
- ∑
- l
- =
- 1
- L
-
- ∑
- i
- ,
- j
- ∈
- 𝒜
- i
- ≠
- j
-
- L
- -
- l
- L
- ·
- A
-
- (
- i
- ,
- j
- )
- ·
- P
-
- (
- l
- ,
- i
- ,
- j
- )
- ·
-
- v
- i
- l
- -
- v
- j
- l
-
- 2
- where P(l, i, j)=Puse(l, i)·Puse(l, j) is the probability that the nodes i and j in layer l are used in the sampled tree structure, and A(i, j) captures the task affinity between tasks, τi and τj, and
- υ
- i
- l
- and
- υ
- j
- l
- are samples, represents active tasks and L represents the total number of layers;
- training a weight hypernet, keeping the anchor net and the edge hypernet fixed;
- and
- generating, by the hypernetwork before inference time, an architectural configuration for a main model based on the task preference vector and the resource constraint.
A non-transitory computer program product for building a dynamic multi-task network, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform a method comprising:
- generating a hypernetwork configured to model tradeoffs among a plurality of competing tasks;
- receiving a task preference vector identifying a hierarchical priority for the plurality of competing tasks, and a resource constraint as a tuple;
- finding tree sub-structures and the corresponding modulation of features for every tuple within an N-stream anchor network;
- optimizing a branching regularized loss function to train an edge hypernet, wherein the branching regularized loss function is based on an active loss Lactive and an inactive loss, the active loss, Lactive, is calculated as:
- ℒ
- active
- =
- ∑
- l
- =
- 1
- L
-
- ∑
- i
- ,
- j
- ∈
- 𝒜
- i
- ≠
- j
-
- L
- -
- l
- L
- ·
- A
-
- (
- i
- ,
- j
- )
- ·
- P
-
- (
- l
- ,
- i
- ,
- j
- )
- ·
-
- v
- i
- l
- -
- v
- j
- l
-
- 2
- where P(l, i, j)=Puse(l, i)·Puse(l,j) is the probability that the nodes i and j in layer l are used in the sampled tree structure, and A(i,j) captures the task affinity between tasks, τi and τj, and
- υ
- i
- l
- and
- υ
- j
- l
- are samples, represents active tasks and L represents the total number of layers;
- calculating an active loss and an inactive loss to determine branching within the anchor network
- training a weight hypernet, keeping the anchor network and the edge hypernet fixed;
- and
- generating, by the hypernetwork before inference time, an architectural configuration for a main model based on the task preference vector and the resource constraint.
Dependencies are inferred from the claim text.