Detecting and responding to malicious acts directed towards machine learning models
A system detects and responds to malicious acts directed towards machine learning models. Data fed into and output by a machine learning model is collected by a sensor. The data fed into the model includes vectorization data, which is generated from raw data provided from a requester, such as for example a stream of timeseries data. The output data may include a prediction or other output generated by the machine learning model in response to receiving the vectorization data. The vectorization data and machine learning model output data are processed to determine whether the machine learning model is being subject to a malicious act (e.g., attack). The output of the processing may indicate an attack score. A response for handling the request by a requester may be selected based on the output that includes the attack score, and the response may be applied to the requestor.
1. A method for monitoring a machine learning-based system for malicious acts comprising:
receiving vectorization data by a sensor being executed on a server, the vectorization data derived from input data intended for a first machine learning model and provided by a requestor;
receiving, by the sensor, an output generated by the machine learning model, the machine learning model generating the output in response to receiving the vectorization data;
transmitting vectorization data and the output to a processing engine by the sensor;
processing the vectorization data and the output by the processing engine to generate an attack score, the attack score indicating a likelihood of a malicious action towards the machine learning model via the vectorization data; and
applying a response to a request associated with the requestor, the response based at least in part on the attack score, the response applied in place of the output of the first machine learning model, the applying of the response including:
selecting, by a response engine, a response based on an output by a second machine learning model within the processing engine, the output of the second machine learning model including a prediction of an attack on the first machine learning model; and
causing the output of the first machine learning model to be modified or causing the requestor to be disconnected.
2. The method of claim 1 , further comprising collecting the vectorization data by a sensor component, the sensor component transmitting the collected vectorization data to the processing engine on the server.
3. The method of claim 2 , wherein the sensor component is created in a computing environment that proxies the first machine learning model.
4. The method of claim 2 , further including:
collecting the output generated by the first machine learning model by the sensor component;
coupling the vectorization data and output by the sensor component; and
transmitting the coupled vectorization data and output to the processing engine by the sensor component.
5. The method of claim 2 , further including:
intercepting the output of the first machine learning model by a sensor component; and
transmitting a response generated by the sensor to the requestor in place of the output, the response generated based at least in part on the attack score.
6. The method of claim 1 , further comprising generating an alert based on the attack score.
7. The method of claim 1 , further comprising reporting attack data to a user through a graphical interface, the attack data based at least in part on the attack score.
8. A non-transitory computer readable storage medium having embodied thereon a program, the program being executable by a processor to perform operations for monitoring a machine learning-based system for malicious acts, the operations comprising:
receiving vectorization data by a sensor, the vectorization data derived from input data intended for a first machine learning model and provided by a requestor;
receiving, by the sensor, an output generated by the machine learning model, the machine learning model generating the output in response to receiving the vectorization data;
transmitting the vectorization data and the output to a processing engine by the sensor;
processing the vectorization data and the output by the processing engine to generate an attack score, the attack score indicating a likelihood of a malicious action towards the machine learning model via the vectorization data; and
applying a response to a request associated with the requestor, the response based at least in part on the attack score, the response applied in place of the output of the first machine learning model, the applying of the response including:
selecting, by a response engine, a response based on an output by a second machine learning model within the processing engine, the output of the second machine learning model including a prediction of an attack on the first machine learning model; and
causing the output of the first machine learning model to be modified or causing the requestor to be disconnected.
9. The non-transitory computer readable storage medium of claim 8 , wherein the operations further comprise: collecting the vectorization data by a sensor component, the sensor component transmitting the collected vectorization data to the processing engine.
10. The non-transitory computer readable storage medium of claim 9 , wherein the sensor component is created in a computing environment that implements the first machine learning model.
11. The non-transitory computer readable storage medium of claim 9 , wherein the operations further comprise:
collecting the output generated by the first machine learning model by the sensor component;
coupling the vectorization data and output by the sensor component; and
transmitting the coupled vectorization data and output to the processing engine by the sensor component.
12. The non-transitory computer readable storage medium of claim 9 , wherein the operations further comprise:
intercepting the output of the first machine learning model by a sensor component; and
transmitting a response generated by the sensor to the requestor in place of the output, the response generated based at least in part on the attack score.
13. The non-transitory computer readable storage medium of claim 8 , wherein the operations further comprise: generating an alert based on the attack score.
14. The non-transitory computer readable storage medium of claim 8 , wherein the operations further comprise: reporting attack data to a user through a graphical interface, the attack data based at least in part on the attack score.
15. A system for monitoring a machine learning-based system for malicious acts, comprising:
at least one data processor; and
memory for storing instructions which, when executed by the at least one data processor, result in operations comprising:
receiving vectorization data derived from input data intended for a first machine learning model and provided by a requestor;
receiving an output generated by the machine learning model, the machine learning model generating the output in response to receiving the vectorization data;
transmitting vectorization data and the output to a processing engine;
processing the vectorization data and the output by the processing engine to generate an attack score, the attack score indicating a likelihood of a malicious action towards the machine learning model via the vectorization data; and
causing a response to a request associated with the requestor to be applied, the response based at least in part on the attack score, the response applied in place of the output of the first machine learning model, the causing including:
selecting, by a response engine, a response based on an output by a second machine learning model within the processing engine, the output of the second machine learning model including a prediction of an attack on the first machine learning model; and
causing the output of the first machine learning model to be modified or causing the requestor to be disconnected.
16. The system of claim 15 , wherein the operations further comprise: collecting the vectorization data by a sensor component, the sensor component transmitting the collected vectorization data to the processing engine.
17. The system of claim 16 , wherein the sensor component is created in a computing environment that implements the first machine learning model.
18. The system of claim 16 , wherein the operations further comprise:
collecting the output generated by the first machine learning model by the sensor component;
coupling the vectorization data and output by the sensor component; and
transmitting the coupled vectorization data and output to the processing engine by the sensor component.
19. The system of claim 15 , wherein the operations further comprise:
intercepting the output of the first machine learning model by a sensor component; and
transmitting a response generated by the sensor to the requestor in place of the output, the response generated based at least in part on the attack score.
20. The system of claim 15 , wherein the operations further comprise:
generating an alert based on the attack score.