Intelligent task offloading
A method of operating a device that interacts with a physical environment includes determining to take an action on the physical environment wherein the action is based on an output of a computing task, generating a risk level that estimates a risk of physical harm associated with taking the action within the physical environment, obtaining a performance indicator of a communication network between the device and a remote computing device, and determining, based on the risk level and the performance indicator, whether to offload the computing task to the remote computing device or to perform the computing task locally. Related devices are also disclosed.
1. A method of operating a device that interacts with a physical environment, the method comprising:
determining to take an action on the physical environment, wherein the action is based on an output of a computing task;
generating a risk level that estimates a risk of physical harm associated with taking the action within the physical environment;
obtaining a performance indicator of a communication network between the device and a remote computing device;
determining, based on the risk level and the performance indicator, whether to offload the computing task to the remote computing device or to perform the computing task locally,
wherein the deciding whether to offload the computing task to the remote computing device or to perform the computing task locally is performed using a reinforcement learning agent that evaluates a reward for offloading the computing task to the remote device, wherein the reward is calculated based on a latency reward r l , that is based on an amount of time needed to perform the computing task, an energy reward r e that is based on an amount of energy needed to perform the computing task, and an accuracy reward r a that is based on an accuracy of the computing task, and
wherein the latency reward r l is defined as:
r l =w l ×1/ L
where w l is a weight that indicates a relative importance of the latency reward compared to other reward factors and L is a total latency associated with performing the computing task, including a communication latency and an execution latency; and
in response to determining to offload the computing task to the remote computing device, transmitting task input data to the remote computing device, receiving task output data from the remote computing device, and taking the action on the physical environment based on the task output data received from the remote computing device.
2. The method of claim 1 , further comprising:
obtaining a set of task input data for performing the computing task;
comparing the task input data for performing the computing task with a previous set of task input data for performing the same computing task; and
deciding to use a set of task output data that was generated based on the previous set of task input data instead of offloading the computing task to the remote device or performing the computing task locally.
3. The method of claim 1 , wherein the performance indicator of the communication network comprises at least one of a throughput, a round-trip-time, a bandwidth and a latency.
4. The method of claim 1 , wherein the communication latency comprises a time required to transmit a set of task input data to a task processing module that will perform the computing task and to receive a set of output data from the task processing module, and wherein the execution latency comprises a time required for the task processing module to complete the computing task.
5. A method of operating a device that interacts with a physical environment, the method comprising:
determining to take an action on the physical environment, wherein the action is based on an output of a computing task;
generating a risk level that estimates a risk of physical harm associated with taking the action within the physical environment;
obtaining a performance indicator of a communication network between the device and a remote computing device;
determining, based on the risk level and the performance indicator, whether to offload the computing task to the remote computing device or to perform the computing task locally,
wherein the deciding whether to offload the computing task to the remote computing device or to perform the computing task locally is performed using a reinforcement learning agent that evaluates a reward for offloading the computing task to the remote device, wherein the reward is calculated based on a latency reward r l , that is based on an amount of time needed to perform the computing task, an energy reward r e that is based on an amount of energy needed to perform the computing task, and an accuracy reward r a that is based on an accuracy of the computing task, and
wherein the energy reward is defined as:
r e =w e ×1/ E
where E is an amount of energy spent to perform the computing task and w e is a weight that indicates a relative importance of the energy reward compared to other reward factors; and
in response to determining to offload the computing task to the remote computing device, transmitting task input data to the remote computing device, receiving task output data from the remote computing device, and taking the action on the physical environment based on the task output data received from the remote computing device.
6. The method of claim 1 , wherein the accuracy reward r a is equal to zero if the computing task is performed locally and is equal to a non-zero number if the computing task is performed by the remote computing device.
7. A method of operating a device that interacts with a physical environment, the method comprising:
determining to take an action on the physical environment, wherein the action is based on an output of a computing task;
generating a risk level that estimates a risk of physical harm associated with taking the action within the physical environment;
obtaining a performance indicator of a communication network between the device and a remote computing device;
determining, based on the risk level and the performance indicator, whether to offload the computing task to the remote computing device or to perform the computing task locally,
wherein the deciding whether to offload the computing task to the remote computing device or to perform the computing task locally is performed using a reinforcement learning agent that evaluates a reward for offloading the computing task to the remote device, wherein the reward is calculated based on a latency reward r l , that is based on an amount of time needed to perform the computing task, an energy reward r e that is based on an amount of energy needed to perform the computing task, and an accuracy reward r a that is based on an accuracy of the computing task, and
wherein the reward is calculated as:
reward=(risk value +1)×( r l +r a )+ r e
where risk value is the risk level; and
in response to determining to offload the computing task to the remote computing device, transmitting task input data to the remote computing device, receiving task output data from the remote computing device, and taking the action on the physical environment based on the task output data received from the remote computing device.
8. The method of claim 1 , wherein the reward is further calculated based on temporal coherence reward that is based on a similarity between a previous task input to a current task input.
9. A method of operating a device that interacts with a physical environment, the method comprising:
determining to take an action on the physical environment, wherein the action is based on an output of a computing task;
generating a risk level that estimates a risk of physical harm associated with taking the action within the physical environment;
obtaining a performance indicator of a communication network between the device and a remote computing device;
determining, based on the risk level and the performance indicator, whether to offload the computing task to the remote computing device or to perform the computing task locally,
wherein the deciding whether to offload the computing task to the remote computing device or to perform the computing task locally is performed using a reinforcement learning agent that evaluates a reward for offloading the computing task to the remote device, wherein the reward is calculated based on a latency reward r l , that is based on an amount of time needed to perform the computing task, an energy reward r e that is based on an amount of energy needed to perform the computing task, and an accuracy reward r a that is based on an accuracy of the computing task, wherein the reward is further calculated based on temporal coherence reward that is based on a similarity between a previous task input to a current task input, and
wherein the temporal coherence reward is calculated as:
r t =(temporal coherence 4 −1)×reward partial .
where temporal coherence is the temporal coherence and reward partial is a reward calculated based on the risk level, the latency reward, the accuracy reward and the energy reward; and
in response to determining to offload the computing task to the remote computing device, transmitting task input data to the remote computing device, receiving task output data from the remote computing device, and taking the action on the physical environment based on the task output data received from the remote computing device.
10. The method of claim 8 , wherein the previous task input and the current task input comprise images of the physical environment.
11. The method of claim 10 , wherein the temporal coherence is calculated based on mutual information of the images of the physical environment.
12. The method of claim 11 , wherein the mutual information of the images of the physical environment is obtained by calculating an entropy of each image and a joint entropy of the images and subtracting the entropy of each image from the joint entropy.
13. The method of claim 1 , wherein generating the risk level comprises:
detecting objects within the physical environment;
generating a semantic representation of the physical environment, wherein the semantic representation of the physical environment includes properties of the detected objects;
classifying the detected objects; and
generating the risk level based on the properties and classifications of the objects.
14. The method of claim 13 , wherein the semantic representation comprises a scene graph.