Okso. I've thought about this question for a while and hemmed and hawed about a way to answer it. I'll start off by keeping the answer brief and then forgetting how to be brief and discussing things.

We don't know. And we have no possibility of knowing.

Okay, now that that's over with, we can talk about the why. Why is it that, when we have ALL THIS INFO about the Roman world, and knowing that Julius Caesar's time period is one of the most well-documented parts of it (thanks, Cicero), answering a reasonably simple query about waste is impossible?

First, and perhaps most importantly, we don't actually know how many people lived in Rome - during this time, or during any time. We have estimates, sure, and we have the various censuses - but the problem is that the censuses themselves are annoyingly difficult to quantify. Who's being counted? Why does the population count jump from 450,000 to 4,000,000 within about 60 years? Are any of these numbers even accurate, and how accurate should we treat them? (The current "consensus," for the most part, assumes that the 4,000,000 number counts all Roman men and women within Italy. But then, of course, we have to ask how accurate this number actually is...) The problem here, as with so many things in antiquity, is a frustrating lack of real data. We have a fraction of a fraction of a single percent of documents that survive in some way, shape, or form - and an enormous number of those have been copied and recopied a few times over (which is problematic for every reason that a game of "telephone" is problematic, but that's an entirely different topic).

So the census data is difficult to use at best. Haven't classicists and ancient historians found out ways to estimate populations, though?

Well, yes and no. The biggest problem is that Classics is the definition of the Ivory Tower discipline. It doesn't develop well with the times, and as such, often eschews modern research. The first real, new studies about Roman demography only started coming out relatively recently (within the last few decades or so). Oftentimes, it's easier to just lean on the crutch of philology, note that ancient authours put Rome's population at a million, dust off your hands and move on to the next book. Modern demographers love arguing with that number, and all I can say with confidence is that Rome was a big city. We just don't know how big. And because we don't know how big the city was, it's tough to estimate how much food the city needed to survive (discounting the conversion rates of "grain" into usable food, other supplemental foods, food that was sold through street vendors and possibly incidentally wasted, food that simply spoiled due to a lack of refrigeration, etc), let alone waste. We have the epic trash heaps around the empire (behold the 8th hill of Rome), but those only tell us about the incredible amount of importing that the city did. Which...is incredible, but also not particularly helpful to the question at hand.

So we don't know and can't easily estimate the population because of a lack of records. What other problems are there?

Well, slaves. How many were there? We have no idea. We know that slavery was endemic, hugely widespread, and woven into every facet of Roman life. We don't know how many slaves there were, we don't know the actual ratio of field slaves to city slaves, we don't know how many of them became libertus/a, and of those, we still can't tell much because they still wouldn't be on the census. We've got their gravestones (well, some of them), which is nice, but still incredibly unhelpful for the most part, because that still doesn't tell us how many of them there were, how much each slave was allowed to consume, and whether or not a slave had enough leeway to be wasteful (Okay, that last bit is probably city slaves only, for the most part). But that brings us to another question - what counts as wasted food? Does grain that isn't harvested enough count? Does accidentally trampling an apple? How about a potentially preventable pest that ruins a crop?

These are all potentially silly questions, but they're incredibly important for the question at hand (and hell, I'd love to know even just one of the answers, myself).

So all of the big, overarching data is bad. How about the little stuff? How much would an "average Roman" waste during this time period?

Well, it all goes back (again) to a lack of solid data, in that we just don't know. We know that Rome was a reasonably filthy place, that's always a fun topic. We know that they had take out (or their version of it, where you could go and get your Subway sandwich made and bring it home to nosh on). We also know that the grain dole was a part of life, and an enormous portion of the (also enormous) city relied on the free (or HEAVILY discounted) grain to keep themselves alive. These poor would not necessarily have anything to waste. The elites, on the other hand, could be incredibly wasteful, as beautifully illustrated in the surviving fragments of the Satyricon - more specifically, the infamous Cena Trimalchionis (starts here and goes on for a bit). But while we have plenty of complaints about the wastefulness of the elite (Stoic philosophers love to do this, as do biographers who hate the emperor they're talking about), those are the 1 % of the 1%, and cannot be taken as an average number, nor even entirely accurate accounts, depending.

And on top of that, even if we DID extrapolate from all evidence, even if we DID estimate how much food "An Elite" wasted, we don't actually know how many of them there were. Sure, there was a certain number of senators. But how about their wives? Their children? Their relatives? How about the equestrians, especially those who were ridiculously wealthy themselves?

I'm sorry I can't give you a better answer than this - the problem is that you've posited an unanswerable question. There have been multiple dissertations written attempting to grapple with single parts of this problem, and none of them have been exceptionally satisfactory. But I hate having to leave you at one upvote with no answers or explanations as to why this question can't be answered. So I hope that I helped you to see a couple of the problems that Classicists are currently grappling with :)