Logic and Proof supervisions
This course has three supervisions over the Lent term. I recommend you and your partner book all of your supervisions NOW — this will help to prevent supervisions bleeding into your well-earned holidays:) (If none of the available times on KuDoS work for you, please let me know.)
I prefer submitted work to be LaTeXed. If you’re finding it difficult to lay out proof trees, here’s a LaTeX package for proof-building to avoid the nested frac{}{}
malarkey (link).
Supervision 1
- Parts (b) and (c) of 2017P6Q6 (link)
- Part (b) of 2015P6Q6 (link)
- Parts (a) and (b) of 2010P6Q6 (link)
- I am trying to prove safety properties of digital circuits with a theorem prover using propositional logic. What are the practical advantages and disadvantages of using first-order logic for this application instead? (Hint: what’s the domain?)
Supervision 2
- Part (a) of 2019P6Q9 (link)
- 2016P6Q5 (link)
- Assuming the three statements (a) \(H \rightarrow M \vee N\), (b) \(M \rightarrow K \wedge P\), and (c) \(N \rightarrow L \wedge P\), prove \(H \rightarrow P\) using clause resolution.
- 2020P6Q9 (link)
- (Optional) — From the notes, DPLL has five steps. For each, justify why it is necessary for the algorithm to work.
Supervision 3
- Part (b) of 2019P6Q10. For (i) and (ii) use tableau calculus, for (iii) use sequent calculus. (link)
- Part (b) of 2019P6Q9 (link)
- Part (d) of 2021P6Q10 (link)
- 2016P6Q6 (link)
If you’re having trouble understanding BDDs: link