UiPath Specialized AI Professional
Questions and Answers
Specialized AI - Correct Answers -AI which has been trained to perform a specific task.
UiPath Communications Mining - Correct Answers -Uses a proprietary LLM with NLP
capabilities. Was built and trained specifically for short-form asynchronous data, it can
be considered Specialized AI.
UiPath Document Understanding - Correct Answers -Utilizes both Generative and
Specialized AI to classify attachments and extract relevant info from them.
Tabs on AI Center Dashboard - Correct Answers -Datasets
- Upload datasets
Data Labeling
- Create "data labeling sessions??"
ML Packages
- Upload custom model created w/ Python
- Import ML Package (upload model previously exported from AI Center)
- Out of the box packages (Premade ones by UiPath and Open Source).
Pipelines
- Use dataset and ML package together to produce an output --> output depends on
pipeline type.
ML Skills
ML Logs
- Any info related to ML packages i logged here
- Pipeline start, etc.
Pipelines in AI Center - Correct Answers -Full Pipeline
- Training + Evaluation pipeline
Train Run
- Takes dataset --> Outputs new model version
,Evaluation Run
- Takes dataset --> Metrics + logs
Pipeline - Correct Answers -The description of an ML workflow, including the input data
and the ML package.
ML Package - Correct Answers -A folder with all the code and metadata needed to train
and serve an ML model.
ML Skills Public Mode - Correct Answers -Mode allows for robot to not be connected to
the same tenant that the ML skill is running in.
ML Skill connection modes - Correct Answers -Robot
- Robot must be connected to the same tenant
Endpoint
- For public ML skills --> Hit it with an API
Task Mining - Correct Answers -Records and captures the tasks performed directly on a
user's desktop and visualizes their workflow to identify repetitive activities and variations
- prime candidates for automation.
Red Dial Indicator - Correct Answers -Appears when a model has fewer than 25
examples.
Comms. Mining - Alert Centre - Correct Answers -Houses the issues and alerts page.
Alerts trigger when certain criteria are met, and these are tracked in the issues page.
Volume Based Alerts - Correct Answers -Focus on communications volume monitoring.
For example, number of exceptions or errors, or any instance of a high risk event.
Active Pipelines - Correct Answers -Pipelines in a project that are currently running, or
scheduled.
Permissions at Project Level - Correct Answers -Permissions
- Datasets
- Pipelines
- Labeling
- ML Packages
- ML Skills
- ML Logs (read only)
- Settings: User management at project level
Project Creator - Correct Answers -Added to the person who created the project.
Administrators Group (Suite -> AI Center) - Correct Answers -Is added with the Users
Manager role.
, Everyone Group (Suite -> AI Center) - Correct Answers -Is added with the Data
Scientist role.
Project (AI Center) - Correct Answers -An isolated group of resources, consisting of
Datasets, ML Packages, Pipelines, ML Skill, and ML Logs.
AI Center Role Based Access Control - Two Levels of Management - Correct Answers -
1. Project
* Datasets
* Pipelines
* Data Labeling
* ML Packages
* ML Skills
* ML Logs
* Settings: User Management at Project Level
2. Tenant
UiPath Document Manager - Correct Answers -- Import and manage documents for
labeling
- Define and configure fields to be extracted
- Pre-label documents using an existing model
- Export documents in the format expected by the Doc. Understanding models
Confidence - Correct Answers -A numerical value that is meant to output an algorithm's
estimation of how well it has performed a certain task.
Enable Infra Settings - Correct Answers -Toggle to customize the resources and
storage size for the specific deployment (ONLY ON-PREM)
* Request and limit size for RAM (GB)
* The request and limit size for CPU
* Storage size (GB)
Scheduled - Correct Answers -ML Pipeline Status -
A pipeline that has been scheduled to start in the future (for example at 1am every
Monday). When the date-time set for a pipeline to start running is reached, the pipeline
is queued to run.
Waiting For Resources - Correct Answers -ML Pipeline Status -
Waiting for resources because there is no license available for the pipeline to execute.
Every 5 minutes, pipeline checks for an available license to execute (this will happen if
you remove running ML skills or if another pipeline is complete) and starts as soon as
this is the case.)
Questions and Answers
Specialized AI - Correct Answers -AI which has been trained to perform a specific task.
UiPath Communications Mining - Correct Answers -Uses a proprietary LLM with NLP
capabilities. Was built and trained specifically for short-form asynchronous data, it can
be considered Specialized AI.
UiPath Document Understanding - Correct Answers -Utilizes both Generative and
Specialized AI to classify attachments and extract relevant info from them.
Tabs on AI Center Dashboard - Correct Answers -Datasets
- Upload datasets
Data Labeling
- Create "data labeling sessions??"
ML Packages
- Upload custom model created w/ Python
- Import ML Package (upload model previously exported from AI Center)
- Out of the box packages (Premade ones by UiPath and Open Source).
Pipelines
- Use dataset and ML package together to produce an output --> output depends on
pipeline type.
ML Skills
ML Logs
- Any info related to ML packages i logged here
- Pipeline start, etc.
Pipelines in AI Center - Correct Answers -Full Pipeline
- Training + Evaluation pipeline
Train Run
- Takes dataset --> Outputs new model version
,Evaluation Run
- Takes dataset --> Metrics + logs
Pipeline - Correct Answers -The description of an ML workflow, including the input data
and the ML package.
ML Package - Correct Answers -A folder with all the code and metadata needed to train
and serve an ML model.
ML Skills Public Mode - Correct Answers -Mode allows for robot to not be connected to
the same tenant that the ML skill is running in.
ML Skill connection modes - Correct Answers -Robot
- Robot must be connected to the same tenant
Endpoint
- For public ML skills --> Hit it with an API
Task Mining - Correct Answers -Records and captures the tasks performed directly on a
user's desktop and visualizes their workflow to identify repetitive activities and variations
- prime candidates for automation.
Red Dial Indicator - Correct Answers -Appears when a model has fewer than 25
examples.
Comms. Mining - Alert Centre - Correct Answers -Houses the issues and alerts page.
Alerts trigger when certain criteria are met, and these are tracked in the issues page.
Volume Based Alerts - Correct Answers -Focus on communications volume monitoring.
For example, number of exceptions or errors, or any instance of a high risk event.
Active Pipelines - Correct Answers -Pipelines in a project that are currently running, or
scheduled.
Permissions at Project Level - Correct Answers -Permissions
- Datasets
- Pipelines
- Labeling
- ML Packages
- ML Skills
- ML Logs (read only)
- Settings: User management at project level
Project Creator - Correct Answers -Added to the person who created the project.
Administrators Group (Suite -> AI Center) - Correct Answers -Is added with the Users
Manager role.
, Everyone Group (Suite -> AI Center) - Correct Answers -Is added with the Data
Scientist role.
Project (AI Center) - Correct Answers -An isolated group of resources, consisting of
Datasets, ML Packages, Pipelines, ML Skill, and ML Logs.
AI Center Role Based Access Control - Two Levels of Management - Correct Answers -
1. Project
* Datasets
* Pipelines
* Data Labeling
* ML Packages
* ML Skills
* ML Logs
* Settings: User Management at Project Level
2. Tenant
UiPath Document Manager - Correct Answers -- Import and manage documents for
labeling
- Define and configure fields to be extracted
- Pre-label documents using an existing model
- Export documents in the format expected by the Doc. Understanding models
Confidence - Correct Answers -A numerical value that is meant to output an algorithm's
estimation of how well it has performed a certain task.
Enable Infra Settings - Correct Answers -Toggle to customize the resources and
storage size for the specific deployment (ONLY ON-PREM)
* Request and limit size for RAM (GB)
* The request and limit size for CPU
* Storage size (GB)
Scheduled - Correct Answers -ML Pipeline Status -
A pipeline that has been scheduled to start in the future (for example at 1am every
Monday). When the date-time set for a pipeline to start running is reached, the pipeline
is queued to run.
Waiting For Resources - Correct Answers -ML Pipeline Status -
Waiting for resources because there is no license available for the pipeline to execute.
Every 5 minutes, pipeline checks for an available license to execute (this will happen if
you remove running ML skills or if another pipeline is complete) and starts as soon as
this is the case.)
