Symbolic mathematical calculations in Clojure vs. F# - f#

I have come across the following F# sample and found it intriguing.
http://www.codeproject.com/KB/net-languages/SymbolicCalcInFS.aspx
Does Clojure have language/library facilities for doing something like this with ease? It is ok to force Polish notation for formulas, if that makes things easier.
Thanks, and let me know if there are questions.

Lisp has long history in symbolic computing. See the AI case study book by Peter Norvig. Lisp provides a lot of great language features to abstract the common operations on the symbols. Sometimes you can write really concise code (more concise/short than F#).
Static languages like F# have strong type systems and convenient pattern matching on the data types. The compiler could find errors that are caught by the type system, e.g. lack of considering one special case. Thinking about types with your data could also reduce the chances for runtime error. The type inference in F# also makes F# code very concise.

I don't know much about Clojure, but here are, at least, some pointers.
The key feature that makes the F# code nice is pattern matching on algebraic data types. An algebraic data type is for example the declaration of the Expression type (that is used to represent mathematical expressions) and pattern matching is the match construct that is used to check for various known cases when implementing simplification or differentiation.
I don't think that Clojure has any built-in support for pattern matching, but it can be implemented as a library. One library that looks quite interesting is the patter-match module (in Clojars). Here is an example that uses it to implement algebraic evaluator (which is quite close to the F# article).
Another thing that appears in the F# article is active patterns (which allow you to declare and reuse patterns). I don't think that there is a Clojure library for that, but given the flexibility of the language, it should be possible to implement them too (however, they are not really that necessary in the F# article)

Symbolic differentiation was one of the first applications of lisp!
I made a blogpost about a simple symbolic differentiator. It only deals with + and *, but it is easily extended.
It was part of a series I wrote to introduce beginners to clojure at a conference in London, to show how easy it is for clojure to manipulate its own code.
Of course the cute thing is that having done the differentiation, the code can then be compiled! So you can produce differentiated versions of user input, or macros that produce functions and their derivatives, etc.
The original's here, and nicely syntax highlighted:
http://www.learningclojure.com/2010/02/clojure-dojo-4-symbolic-differentiation.html
But I've posted the code here so you can have a look:
;; The simplest possible symbolic differentiator
;; Functions to create and unpack additions like (+ 1 2)
(defn make-add [ a b ] (list '+ a b))
(defn addition? [x] (and (=(count x) 3) (= (first x) '+)))
(defn add1 [x] (second x))
(defn add2 [x] (second (rest x)))
;; Similar for multiplications (* 1 2)
(defn make-mul [ a b ] (list '* a b))
(defn multiplication? [x] (and (=(count x) 3) (= (first x) '*)))
(defn mul1 [x] (second x))
(defn mul2 [x] (second (rest x)))
;; Differentiation.
(defn deriv [exp var]
(cond (number? exp) 0 ;; d/dx c -> 0
(symbol? exp) (if (= exp var) 1 0) ;; d/dx x -> 1, d/dx y -> 0
(addition? exp) (make-add (deriv (add1 exp) var) (deriv (add2 exp) var)) ;; d/dx a+b -> d/dx a + d/dx b
(multiplication? exp) (make-add (make-mul (deriv (mul1 exp) var) (mul2 exp)) ;; d/dx a*b -> d/dx a * b + a * d/dx b
(make-mul (mul1 exp) (deriv (mul2 exp) var)))
:else :error))
;;an example of use: create the function x -> x^3 + 2x^2 + 1 and its derivative
(def poly '(+ (+ (* x (* x x)) (* 2 (* x x))) 1))
(defn poly->fnform [poly] (list 'fn '[x] poly))
(def polyfn (eval (poly->fnform poly)))
(def dpolyfn (eval (poly->fnform (deriv poly 'x))))
;;tests
(use 'clojure.test)
(deftest deriv-test
(testing "binary operators"
(is (= (let [m '(* a b)] [(multiplication? m) (make-mul (mul1 m) (mul2 m))]) [true '(* a b)]))
(is (= (let [m '(* a b)] [(addition? m) (make-add (add1 m) (add2 m))]) [false '(+ a b)])))
(testing "derivative function"
(is (= (deriv '0 'x) '0))
(is (= (deriv '1 'x) '0))
(is (= (deriv 'x 'x) '1))
(is (= (deriv 'y 'x) '0))
(is (= (deriv '(+ x x) 'x) '(+ 1 1)))
(is (= (deriv '(* x x) 'x) '(+ (* 1 x) (* x 1))))
(is (= (deriv '(* x x) 'y) '(+ (* 0 x) (* x 0))))
(is (= (deriv '(* x (* x x)) 'x) '(+ (* 1 (* x x)) (* x (+ (* 1 x) (* x 1)))))))
(testing "function creation: d/dx (x^3 + 2x^2 + 1) = 3x^2 + 4x "
(let [poly '(+ (+ (* x (* x x)) (* 2 (* x x))) 1)]
(is (= ((eval (poly->fnform poly)) 3) 46))
(is (= ((eval (poly->fnform (deriv poly 'x))) 3))))))

I haven't tried it, but Clojuratica looks very interesting.

Well, now Clojure offer powerful pattern matching libraries :
matchure : https://github.com/dcolthorp/matchure
match : https://github.com/swannodette/match

Yes, a system such as you describe now exists on Clojure! It is none other than Gerry Sussman's companion system to his -and Wisdom's- SICM (Structure and Interpretation of Classical Mechanics) book. For Clojure it's been named sicmutils, and ported by Colin Smith.
I've described it briefly elsewhere - https://stackoverflow.com/a/41646455/4070712 - but in short yes, it definitely does the four things you F# article mentions, viz.
Differentiation:
Simplification of algebraic expressions
Formatting
Parsing of expressions
and much, much more...
1) Differentiation (full partial differentiation supported)
> (defn ff [x y] (* (expt x 3)(expt y 5)))
> ((D ff) 'x 'y) ==> (down (* 3 (expt x 2) (expt y 5)) (* 5 (expt x 3) (expt y 4)))
> ;; i.e. vector of results wrt to both variables
NB. Two types of vectors are supported, "up" and "down" to accomodate covariant and contravariant expressions
2) Simplification of expressions: Oh, yes...
> (def unity (+ (square sin) (square cos)))
> (unity 'x) ==> 1 ;; yes we can deal with symbols
3) Formatting: Expressions can be rendered in TeX for beautiful display.
I can't show this easily here, but currently a Maple-style notebook/workshhet is under development, using Clojure's "Gorilla"
4) Parsing: Obviously. Converting between expressions and functions is a core part of the system.
Have a look at https://github.com/littleredcomputer/sicmutils . you don't even need Clojure to run it, you can use the supplied Java jar file.

Related

Creating a transitive and not reflexive function in Z3

I'm trying to create a function in Z3 that is transitive but not reflexive. I.e. if (transitive a b) and (transitive b c)hold then (transitive a c) should hold, but (transitive a a) should not.
I've tried to do it the following way, with 5 "tests". The first does what I expect, but the second one fails and results in unknown.
(declare-datatypes () ((T T1 T2 T3)))
(declare-fun f (T T) Bool)
(assert(f T1 T2))
(assert(f T2 T3))
; Make sure that f is not reflexive
(assert
(forall ((x T))
(not (f x x))))
; Now we create the transitivity function ourselves
(define-fun-rec transitive ((x T) (y T)) Bool
(or
(f x y)
(exists ((z T))
(and
(f x z)
(transitive z y)))))
; This works and gives sat
(push)
(assert (not (transitive T1 T1)))
(assert (not (transitive T2 T2)))
(assert (not (transitive T3 T3)))
(check-sat)
(pop)
; This fails with "unknown" and the verbose flag gives: (smt.mbqi "max instantiations 1000 reached")
(push)
(assert
(forall ((x T))
(not (transitive x x))))
(check-sat)
(pop)
My question is: how does the second test differ from the first? Why does the last one give unknown, whereas the one before that works just fine?
The "verbose" message is a hint here. mbqi stands for model-based-quantifier-instantiation. It's a method of dealing with quantifiers in SMT solving. In the first case, MBQI manages to find a model. But your transitive function is just too complicated for MBQI to handle, and thus it gives up. Increasing the limit will not likely address the problem, nor it's a long term solution.
Short story long, recursive definitions are difficult to deal with, and recursive definitions with quantifiers are even harder. The logic becomes semi-decidable, and you're at the mercy of heuristics. Even if you found a way to make z3 compute a model for this, it would be brittle. These sorts of problems are just not suitable for SMT solving; better use a proper theorem prover like Isabelle, Hol, Coq, Lean. Agda, etc. Almost all these tools offer "tactics" to dispatch subgoals to SMT solvers, so you have the best of both worlds. (Of course you lose full automation, but with quantifiers present, you can't expect any better.)

What is the return statement in Scheme?

I am trying to translate this Python program to Scheme:
def test(x):
if x > 1:
print('foo')
if x > 10:
return
if x == 4:
print('bar')
test(1)
test(2) # prints 'foo'
test(4) # prints 'foo\nbar'
test(11) # prints 'foo'
What is the return statement in Scheme?
In Scheme there isn't an explicit return keyword - it's a lot simpler than that, the value of the last expression in a sequence of expressions is the one that gets returned. For example, your Python code will translate to this, and notice that the (> x 10) case had to be moved to the bottom, so if it's true it can exit the function immediately with a #f value.
(define (test x)
(if (> x 1)
(do-something))
(if (= x 4)
(do-something))
(if (> x 10)
#f))
(test 11)
=> #f
In fact, after reordering the conditions we can remove the last one, but beware: an unspecified value will be returned if x is not 4, according to Guile's documentation - in other words, you should always return a value in each case, and an if expression should have both consequent and alternative parts.
(define (test x)
(if (> x 1)
(do-something))
(if (= x 4)
(do-something)))
(test 11)
=> unspecified
And by the way, I believe the logic in the Python code is a bit off. The first condition will always be evaluated whenever a value of x greater than 1 is passed, but if it is less than or equal to 1 the returned value in Python will be None and in Scheme is unspecified. Also the original function isn't explicitly returning a value - in Python this means that None will be returned, in Scheme the returned value will be either (do-something) if x happens to be 4, or unspecified in any other case.
In Racket the most literal translation is:
#lang racket
(define (test x)
(let/ec return
(when (> x 1)
(do-something))
(when (> x 10)
(return 42))
(when (= x 4)
(do-something))))
(define (do-something)
(display "!"))
(test 11)
The let/ec is short for let/escape-continuation. Look up the equivalent control structure in the manual for your Scheme implementation of choice.
The example displays one ! and then returns 42.
The implicit return of Scheme can be illustrated by comparing how you can implement a simple function, such as square, in Python and scheme.
In Python:
def square(x):
return x*x;
In Scheme:
(define (square x)
(* x x))
As others have said, the last expression's value in a function is its return value, so you just have to arrange for exclusive execution pathways in your code, to achieve this effect.
(if <test> <consequent> <alternative>) is the basic branching operation in Scheme:
(define (test x)
(if (> x 1)
(do_something)
#f)
(if (> x 10)
#f ; return #f
;; else:
(if (= x 4)
(do_something)
;; else:
#f)))
(test 11)
Or we could use cond to avoid the needlessly nested structure in the code:
(define (test x)
(if (> x 1)
(do_something)
#f)
(cond
( (> x 10) #f)
( (= x 4) (do_something))
( else #f)))
You may use call/cc
(define (test x) (call/cc (lambda (k)
(if x
(k x)
(k))
(display "never displayed"))))
> (test 3)
3
> (test #f)
>
You can return no value using (k).
Read about Continuations.
I meet same question as you asked when I tried to implement a "tile" function to duplicate a list for multiple times. The idea is 1) create a variable to store the temporary result; 2) use "append" function in a for-loop; 3) return the temporary result.
The solution seems easy, as follows:
(define (tile ls n)
(define r (list ))
(do ((i 0 (+ i 1)))
((> i (- n 1)))
(begin
(set! r (append r ls))
)
)
r
)
actually a sole variable name itself will let the whole function return the value of this variable. This is because the Scheme always returns the last executed statement as the value of this function.
There is no return statement in Scheme. The body of a procedure is a sequence of expressions and the return value is the value of the last expression. So here is an equivalent Scheme program:
(define (test x)
(if (> x 1)
(display "foo\n"))
(if (not (> x 10))
(if (= x 4)
(display "bar\n"))))
(test 1)
(test 2) ; prints "foo"
(test 4) ; prints "foo\nbar"
(test 11) ; prints "foo"
You can simplify it to this program:
(define (test x)
(if (> x 1)
(display "foo\n"))
(if (= x 4)
(display "bar\n"))))

Adjusting `simplify` tactic in Z3

I've got several questions about Z3 tactics, most of them concern simplify .
I noticed that linear inequalites after applying simplify are often negated.
For example (> x y) is transformed by simplify into (not (<= x y)). Ideally, I would want integer [in]equalities not to be negated, so that (not (<= x y)) is transformed into (<= y x).  I can I ensure such a behavior?
Also, among  <, <=, >, >= it would be desirable to have only one type of inequalities to be used in all integer predicates in the simplified formula, for example <=.   Can this be done?
What does :som parameter of simplify do? I can see the description that says that it is used to put polynomials in som-of-monomials form, but maybe I'm not getting it right.  Could you please give an example of different behavior of simplify with :som set to true and false?
Am I right that after applying simplify arithmetical expressions would always be represented in the form a1*t1+...+an*tn, where ai are constants and ti are distinct terms (variables, uninterpreted constants or function symbols)? In particular is always the case that subtraction operation doesn't appear in the result?
Is there any available description of the ctx-solver-simplify tactic? Superficially, I understand that this is an expensive algorithm because it uses the solver, but it would be interesting to learn more about the underlying algorithm so that I have an idea on how many solver calls I may expect, etc. Maybe you could give a refernce to a paper or give a brief sketch of the algorithm?
Finally, here it was mentioned that a tutorial on how to write tactics inside the Z3 code base might appear. Is there any yet?
Thank you.
Here is an example (with comments) that tries to answer questions 1-4. It is also available online here.
(declare-const x Int)
(declare-const y Int)
;; 1. and 2.
;; The simplifier will map strict inequalities (<, >) into non-strict ones (>=, <=)
;; Example: x < y ===> not x >= y
;; As suggested by you, for integer inequalities, we can also use
;; x < y ==> x <= y - 1
;; This choice was made because it is convenient for solvers implemented in Z3
;; Other normal forms can be used.
;; It is possible to map everything to a single inequality. This is a straightforward modificiation
;; in the Z3 simplifier. The relevant files are src/ast/rewriter/arith_rewriter.* and src/ast/rewriter/poly_rewriter.*
(simplify (<= x y))
(simplify (< x y))
(simplify (>= x y))
(simplify (> x y))
;; 3.
;; :som stands for sum-of-monomials. It is a normal form for polynomials.
;; It is essentially a big sum of products.
;; The simplifier applies distributivity to put a polynomial into this form.
(simplify (<= (* (+ y 2) (+ x 2)) (+ (* y y) 2)))
(simplify (<= (* (+ y 2) (+ x 2)) (+ (* y y) 2)) :som true)
;; Another relevant option is :arith-lhs. It will move all non-constant monomials to the left-hand-side.
(simplify (<= (* (+ y 2) (+ x 2)) (+ (* y y) 2)) :som true :arith-lhs true)
;; 4. Yes, you are correct.
;; The polynomials are encoded using just * and +.
(simplify (- x y))
5) ctx-solver-simplify is implemented in the file src/smt/tactic/ctx-solver-simplify.*
The code is very readable. We can add trace messages to see how it works on particular examples.
6) There is no tutorial yet on how to write tactics. However, the code base has many examples.
The directory src/tactic/core has the basic ones.

Scheme streams with Taylor series

I've been doing some homework, wrote some code and can't actually find the reason why it doesn't work. The main idea of this part of the work is to make a stream that will give me elements of Taylor series of cosine function for a given X (angle i guess). anyways here is my code, I'd be happy if some one could point me to the reasons it doesn't work :)
(define (force exp) exp)
(define (s-car s) (car s))
(define (s-cdr s) (force (cdr s)))
; returns n elements of stream s as a list
(define (stream->list s n)
(if (= n 0)
'()
(cons (s-car s) (stream->list (s-cdr s) (- n 1)))))
; returns the n-th element of stream s
(define stream-ref (lambda (s n)
(if (= n 1)
(s-car s)
(stream-ref (s-cdr s) (- n 1)))))
; well, the name kinda gives it away :) make factorial n!
(define (factorial x)
(cond ((= x 0) 1)
((= x 1) 1)
(else (* x (factorial (- x 1))))))
; this function is actually the equation for the
; n-th element of Taylor series of cosine
(define (tylorElementCosine x)
(lambda (n)
(* (/ (expt -1 n) (factorial (* 2 n))) (expt x (* 2 n)))))
; here i try to make a stream of those Taylor series elements of cosine
(define (cosineStream x)
(define (iter n)
(cons ((tylorElementCosine x) n)
(lambda() ((tylorElementCosine x) (+ n 1)))))
(iter 0))
; this definition should bind cosine
; to the stream of taylor series for cosine 10
(define cosine (cosineStream 10))
(stream->list cosine 10)
; this should printi on screen the list of first 10 elements of the series
However, this doesn't work, and I don't know why.
I'm using Dr.Scheme 4.2.5 with the language set to "Essentials of Programming Languages 3rd ed".
Since I was feeling nice (and nostalgic about scheme) I actually waded through your code to finde the mistakes. From what I can see there are 2 problems which keeps the code from running as it should:
If I understand your code correctly (force exp) should evaluate exp, however you directly return it (unevaluated). So it probably should be defined as (define (force exp) (exp))
The second problem is in your lambda: (lambda() ((tylorElementCosine x) (+ n 1)) ) will evaluate to the next element of the taylor series, while it should evaluate to a stream. You probably want something like this: (lambda() (iter (+ n 1)) )
I haven't checked if the output is correct, but with those modifications it does at least run. So if there are any more problems with the code the should be in the formula used.
However I'd suggest that next time you want help with your homework you at least tell us where exactly the problem manifests and what you tried already (the community does frown on "here is some code, please fix it for me" kind of questions).

Scheme: pattern matching syntax

This was an edit to an earlier post. I am reposting it because I think the original isn't getting any more views since I accepted a partial answer already.
I have written a function match-rewriter which is just match-lambda except that it returns its argument if no match is found.
Using match rewriter I want to be able to write rules that can be passed to another function rewrite which is this:
#| (rewrite rule s) repeatedly calls unary function 'rule' on every "part"
of s-expr s, in unspecified order, replacing each part with result of rule,
until calling rule makes no more changes to any part.
Parts are s, elements of s, and (recursively) parts of the elements of s. (define (rewrite rule s) |#
(let* ([with-subparts-rewritten
(if (list? s) (map (λ (element) (rewrite rule element)) s) s)]
[with-also-rule-self (rule with-subparts-rewritten)])
(if (equal? with-also-rule-self with-subparts-rewritten)
with-also-rule-self
(rewrite rule with-also-rule-self))))
Here is an example of proper usage:
(define arithmetic
(match-rewriter (`(+ ,a ,b) (+ a b))
(`(* ,a ,b) (* a b))
))
(rewrite arithmetic '(+ (* 2 (+ 3 4)) 5))
==>
19
Now I have written:
(define let→λ&call
(match-rewriter (`(let ((,<var> ,<val>) . (,<vars> ,<vals>)) ,<expr> . ,<exprs>)
`((λ (,<var> . ,<vars>) ,<expr> . ,<exprs>) ,<val> . ,<vals>))))
to implement lets as lambda calls, but this is how it is behaving:
(rewrite let→λ&call '(let((x 1) (y 2) (z 3)) (displayln x) (displayln y) (displayln z)))
'((λ (x y 2)
(displayln x)
(displayln y)
(displayln z))
1
z
3)
which, I have to say, really has me stumped. Strangely this call:
(rewrite let→λ&call '(let((w 0) (x 1) (y 2) (z 3)) (displayln w) (displayln x) (displayln y) (displayln z)))
'(let ((w 0) (x 1) (y 2) (z 3))
(displayln w)
(displayln x)
(displayln y)
(displayln z))
Just returns its argument, meaning that match-rewriter did not find a match for this pattern.
Any advice is appreciated.
Thanks.
This pattern:
((,<var> ,<val>) . (,<vars> ,<vals>))
does not do what you want. In particular, it's equivalent to:
((,<var> ,<val>) ,<vars> ,<vals>)
I recommend that you use regular match patterns, rather than quasi-patterns, until you have a better sense of how they work. The pattern for this would be:
(list (list <var> <val>) (list <vars> <vals>) ...)

Resources