thesis-projects issueshttps://gl.kwarc.info/kwarc/thesis-projects/-/issues2017-04-13T08:05:33Zhttps://gl.kwarc.info/kwarc/thesis-projects/-/issues/14semantics extractions based on machine learning2017-04-13T08:05:33ZMichael Kohlhasemichael.kohlhase@fau.desemantics extractions based on machine learningWe have a couple of corpora from which we want to extract semantical features.
Examples are
* quantity expressions like "3m/s" (three meters per second) or "two furlongs per fortnight"
* polarity of identifiers in formulae (essential...We have a couple of corpora from which we want to extract semantical features.
Examples are
* quantity expressions like "3m/s" (three meters per second) or "two furlongs per fortnight"
* polarity of identifiers in formulae (essentially, which symbols in a formula can be substituted for)
* where are "definitions/theorems/assumptions" (and what are their definienda, definienses, and statemnets).
* or more generally what is the content form of a formula
If we know any of those, we could extend nice semantic features (e.g. better screen readers for visually challenged people or better scientific search engines) relatively directly.
We have a couple of large corpora e.g. the [arXMLiv corpus](cortex.mathweb.org/corpus/arXMLiv/tex_to_html) or the data behind the [Online Encyclopaedia of Integer Sequences](http://oeis.org)
All of them are (probably) amenable to machine-learning methods. In some cases, we already have some data about the phenomena above which can act as a baseline.
The topic is to pick one or more of these aspects of semantics and see what contemporary statistical AI methods can do to scale these up to corpus size and develop an symbolic application (possibly with a lot of help from the group).https://gl.kwarc.info/kwarc/thesis-projects/-/issues/34Machine Learning Applications for FrameIT2023-02-03T10:26:41ZRichard MarcusMachine Learning Applications for FrameITIn the FrameIT project (#1), we use the Unity game engine in combination with MMT to have access to mathematical knowledge management techniques and interact with the knowledge in the game world.
Building on the [Unity machine learning ...In the FrameIT project (#1), we use the Unity game engine in combination with MMT to have access to mathematical knowledge management techniques and interact with the knowledge in the game world.
Building on the [Unity machine learning integration](https://unity.com/products/machine-learning-agents), we would like to teach an agent to play the games developed with the FrameIT method.
Possible approaches may be:
- reinforcement learning via reward function
- imitation learning
- modularizing tasks based on the MMT knowledge formalization
Note that there is no groundwork for this topic, so a strong background in machine learning is required.Michael Kohlhasemichael.kohlhase@fau.deMichael Kohlhasemichael.kohlhase@fau.de