Datasets:

kanishka

/

counterfactual-babylm-pipps-random_removal

like 0

Well that's it for our first look at our new building, I hope you're as excited as I am, and I hope to come back in a month or so and give you an updated look at our new building.

MARK MORZE: --people.

What's fascinating about the Z Best case is, number one, we had the real division at Z Best that cleaned carpets.

But accounting work takes a lot of time.

And you can't just say, I need an audit tomorrow.

How could that be?

Watergate hearings, when he'd go, well, I'll just an old, dumb country lawyer.

Yeah.

Grab your wallet and run when somebody says that.

And so what we fraudsters do is we act a little dumb.

Then you basically, by accepting my work at some point, tell me what your benchmark is.

It can be high.

It can be low.

But as long as you tell me what it is, I'll always give you just exactly enough to squeak by.

And in that way, I get you to do the heavy lifting for me.

But if you signed 100 things before, it's just another one, oh, and sign this one, too.

People get suspicious when they think they're trying to be forced into something.

I did hundreds of millions of dollars of transactions using cashier's checks, where I would buy a cashier's check, which the check you use to buy a cashier's check is made out to the bank.

Out of a few hundred million dollars worth of checks that were written, the auditors never saw one canceled cashier's check.

And the bid and acceptance was all the same piece of paper.

Also, it was the front side of one sheet of paper.

like a small phone book.

And I said, yeah, like a million.

He said, could you just write down, during lunch today, some questions that if people had asked, you had no answer for?

Well, I wrote 200 of them in one lunch period.

And each one of those were the type where I really couldn't even dream up a fake answer.

So I sat there in constant fear that someone was going to ask these things.

And lo and behold, no one every did.

No one ever said, how do you pull your permits?

And yet, it was common knowledge.

One auditor is doing your bank reconciliation.

Another one's checking your receivables.

No one ever talk to our vendors because--

Tough Techs, a carpet company.

None of the bogus vendors had any phone numbers.

Their invoices never had phone numbers on them.

No phone calls were ever made to a vendor.

And all of the vendors, supposedly, were within about a three square mile area, right there in the heart of

LA, so they could have-- actually, in the heart of the

I think the biggest lesson that has to be learned is that, if for no other reason than there it doesn't even appear to be a conflict of interest, you must periodically bring in outsiders who have impunity from offending you.

But it's almost always the belief that the real loyalty that any internal auditor, any ethics person, or any compliance officer has is to the company that pays him his check, for god sakes.

There's no favoritism.

He's got a retainer with you.

Remember, they're the real ones that need these kind of protections.

And about the only way you can do that is to bring in someone whose loyalty is, hopefully, to the truth, instead of to keeping the company going.

A lack of due diligence on the people who were hired to do due diligence.

Lastly, I would say to somebody, you have to have the personal courage and integrity to say, I will walk away from this job.

As much as I want my kids in private school, I will walk from this job if this activity continues.

And that is very different from what's been before.

There is no honor among thieves.

And I say, every single day that you're at work, spend two minutes, only two minutes, doing some kind of extra due diligence.

Go online and check industry standards if you don't understand a company that you're doing auditing work for.

Call a bank reference.

Check with Dun and Bradstreet.

Check to see if they're incorporated properly.

See if they pull the right permits for safety violations or whatever it might be.

Visit one of their vendors.

Send an email to somebody that you trust, addressed to them, and then walk away from it.

And you did them just so you could see if there was something hinky or fraudulent or bad out there.

At the end of two or three years, you'll have a volume that you'll be able to show to somebody.

Because in this day and age, and it's an awful thing to have had happen, you frequently have to prove your innocence.

Only you can do that to indemnify yourself.

Then if a fraud happens, if something happens, and a decision maker-- read that as a prosecutor or civil

litigant-- has to go, I'm either going to lean this way on this person or I'm going to lean that way, they will always lean in your favor.

Welcome to this first course in the design and analysis of algorithms.

I'm giving you several different motivations for learning about algorithms.

So, for example, maybe you're given a bunch of numbers and you want to rearrange them into sorted order.

So, why study algorithms?

In, say, computer graphics, you need the computational primitives that are supplied by the study of geometric algorithms.

Let me just state one super obvious one.

You find the, the concepts challenging but perhaps you feel just a tinge smarter after we're done.

So, let's now descend from these lofty generalities and get much more concrete.

So, let's just make that a

In this computational problem, we're given as input, two numbers, let's say with n digits.

Okay, back to integer multiplication.

And then to take each of those n partial products, and add them up.

So, that gives you this first partial product.

22, seven, twelve.

Now, the question I'm interested in is, how much work, how many primitive operations did we do to multiply these two numbers.

Well, just to get sort of a ballpark view for what's going on, we started with two n digit numbers.

And at the end of the day, we basically filled out a grid of size roughly n by n, give or take a little bit.

The grid is n by n roughly so that is roughly n squared operations.

We take the four we multiplied it by each of the n digits on top.

We also did some carries, so that effects things a little bit.

And that's true, of course, for each of the n partial products.

So, that's roughly n squared operations.

Then, we have to add it all up at the end.

And if you want to be an expert algorithm designer, that kind of obedient attitude is something you're going to have to learn to discard.

Here's a favorite quote of mine.

So, this will come up over and over again.

We'll see an algorithm.

What could it possibly look like?

So, as someone with some programming experience, you know that they are not only iterative algorithms, iterative programs, like the one we just outlined for multiplying two integers.

What must such an algorithm look like?

That is, on smaller numbers, numbers with fewer digits.

Well, if you think about it, there's a pretty natural way to break a number with a bunch of digits into a couple numbers with fewer digits.

Namely, just take the first half of the digits, the first n over two digits, regard that as a number in its own right.

Similarly, we will write y, the second input in terms of its first half and second half of digits, c and d.

I'm assuming for convenience here, that n is even.

And then similarly for c and d.