Skip to main content
Tweeted twitter.com/StackCompSci/status/851346024757178368
Clarified the question is about general purpose machine learning.
Link
yters
  • 1.4k
  • 2
  • 13
  • 21

Problem easy for humans and hard for general purpose machine learning, not needing external context?

Explained further what I mean by external context with an example of a problem that humans can solve due to external context.
Source Link
yters
  • 1.4k
  • 2
  • 13
  • 21

Context free problem Problem easy for humans and hard for machine learning, not needing external context?

There are many problems that people can solve more easily than a machine learning algorithm. Often this is attributed to the extra context we have access to.

Image segmentation is an example of a problem that people are good at, but algorithms are not good at, probably because people have lots of external contextual knowledge they bring to bear on the problem. To date, the best techniques use a human in the loop, such as this technique called click carving from HCOMP 2016.

enter image description here

Are there problems that people can solve more easily than most or all machine learning algorithms, where people do not need to depend on access to external context?

Here is oneanother example that is easy for people but hard for a particular algorithm implementation called XGBoost. A human can easily fill in the missing shapes with 100% accuracy, but XGBoost cannot score above 50% accuracy, and its accuracy drops as it gets more training data. Neural networks also seem to have difficulty with this problem.

Note that the person does not need external knowledge about the problem that the algorithm does not have, in order to perform better. All the person needs to see is the configuration of shapes, and they can guess from that what the missing shapes are. The only information the human needs is that the simplest description of the data is most likely the best description, and that the training data is randomly sampled. Both items of information are used in designing XGBoost, and most ML algorithms for that matter.

Easy for humans, hard for XGBoost

Are there other concrete examples like this? Historically, the XOR problem showed that single node perceptrons were inadequate, leading to today's neural network. Besides that, I do not know of any other historical examples.

Context free problem easy for humans and hard for machine learning?

There are many problems that people can solve more easily than a machine learning algorithm. Often this is attributed to the extra context we have access to.

Are there problems that people can solve more easily than most or all machine learning algorithms, where people do not need to depend on access to external context?

Here is one example that is easy for people but hard for a particular algorithm implementation called XGBoost. A human can easily fill in the missing shapes with 100% accuracy, but XGBoost cannot score above 50% accuracy, and its accuracy drops as it gets more training data. Neural networks also seem to have difficulty with this problem.

Easy for humans, hard for XGBoost

Are there other concrete examples like this? Historically, the XOR problem showed that single node perceptrons were inadequate, leading to today's neural network. Besides that, I do not know of any other historical examples.

Problem easy for humans and hard for machine learning, not needing external context?

There are many problems that people can solve more easily than a machine learning algorithm. Often this is attributed to the extra context we have access to.

Image segmentation is an example of a problem that people are good at, but algorithms are not good at, probably because people have lots of external contextual knowledge they bring to bear on the problem. To date, the best techniques use a human in the loop, such as this technique called click carving from HCOMP 2016.

enter image description here

Are there problems that people can solve more easily than most or all machine learning algorithms, where people do not need to depend on access to external context?

Here is another example that is easy for people but hard for a particular algorithm implementation called XGBoost. A human can easily fill in the missing shapes with 100% accuracy, but XGBoost cannot score above 50% accuracy, and its accuracy drops as it gets more training data. Neural networks also seem to have difficulty with this problem.

Note that the person does not need external knowledge about the problem that the algorithm does not have, in order to perform better. All the person needs to see is the configuration of shapes, and they can guess from that what the missing shapes are. The only information the human needs is that the simplest description of the data is most likely the best description, and that the training data is randomly sampled. Both items of information are used in designing XGBoost, and most ML algorithms for that matter.

Easy for humans, hard for XGBoost

Are there other concrete examples like this? Historically, the XOR problem showed that single node perceptrons were inadequate, leading to today's neural network. Besides that, I do not know of any other historical examples.

Source Link
yters
  • 1.4k
  • 2
  • 13
  • 21

Context free problem easy for humans and hard for machine learning?

There are many problems that people can solve more easily than a machine learning algorithm. Often this is attributed to the extra context we have access to.

Are there problems that people can solve more easily than most or all machine learning algorithms, where people do not need to depend on access to external context?

Here is one example that is easy for people but hard for a particular algorithm implementation called XGBoost. A human can easily fill in the missing shapes with 100% accuracy, but XGBoost cannot score above 50% accuracy, and its accuracy drops as it gets more training data. Neural networks also seem to have difficulty with this problem.

Easy for humans, hard for XGBoost

Are there other concrete examples like this? Historically, the XOR problem showed that single node perceptrons were inadequate, leading to today's neural network. Besides that, I do not know of any other historical examples.