Book Review: "Algorithms to Live By" by Brian Christian & Tom Griffiths

I've recently read the book Algorithms to Live By by Brian Christian & Tom Griffiths. This book shows how many problems & heuristics in computer science can be applied to explain or improve human decision-making. Each chapter focuses on a certain class of problems or issues. Such classes include the optimal stopping problem, the multi-armed bandit problem, searching & sorting, task scheduling, Bayesian inference, overfitting data, constraint relaxation, random stimulus, communication protocols, and social interaction. Additionally, most chapters try to show how results from computer science can either improve or justify certain human behaviors.

This book was frustrating for me to read. If it had fully met my expectation that it would show, in a unified & consistent way, how these computer science problems apply to human behavior and connect to each other, I would be singing its praises. If it had completely failed, I'd be happy to rhetorically trash this book. Instead, I found that each chapter would be a great vignette on its own, and each chapter showed the great potential of what the book could have been, but the book failed to live up to that potential. First, there was very little connection among the chapters, and any acknowledgment that the authors did make of such connections was almost always superficial instead of deeply insightful. For example, the respective chapters about the optimal stopping problem, caches, and overfitting each could have been so much better with greater discussion about the connection to social pressure & game theory, yet those topics were discussed only in the last chapter, which I think was a mistake. Second, only in the concluding section did the authors make clear that they wanted to either improve or justify human behavior with each class of problems or issues. This because clear over the course of reading the book, yet there was very little guidance in each chapter about whether improvement versus justification would be the goal. Perhaps the worst offender was the chapter about constraint relaxation, as there was little connection to human behavior in a way that would be obvious to lay readers. These problems meant that reading the last numbered chapter (about game theory) and the conclusion felt simultaneously wonderful for finally seeing these concepts discussed clearly and maddening for knowing that the book could have been so much better if these ideas had been more consistently executed through the book.

There are two other minor criticisms I have of the book too. First, the chapter about overfitting seems to use the word "overfitting" to mean too many different things, which is ironic and undermines any clarity that the discussion could have provided. Second, the chapter about randomized algorithms attempts to make a tenuous connection between randomized algorithms used in computer science and the way that random mental stimuli can produce very creative responses in people, but it never makes clear whether the latter result is true at an individual level or only holds statistically for large populations.

Overall, I think the author's goals were laudable and that each chapter is interesting to read in isolation. However, other readers may be disappointed, as I was, in the way that the authors fail to synthesize many of the ideas across chapters in a smooth & unified manner. Thus, I would advise that readers who may be interested in these topics go into this book with lower expectations.

FOLLOW-UP: How to Tell Whether a Functional is Extremized

This post is a follow-up to an earlier post (link here) about how to tell whether a stationary point of a functional is a maximum, minimum, or saddle point. In particular, as I thought about it more, I realized that using the analogy to discrete vectors could help when formulating a more general expression for the second derivative of the nonrelativistic classical action for a single degree of freedom (i.e. the corresponding Hessian operator). Additionally, I thought of a few other examples of actions whose Hessian operators are positive-definite. Finally, I've thought more about how to express these equations for systems with multiple degrees of freedom (DOFs) as well as for fields and about how these ideas connect to the path integral formulation of quantum mechanics. Follow the jump to see more

How to Tell Whether a Functional is Extremized

I happened to be thinking recently about how to tell when a functional is extremized. Examples in physics include minimizing the ground state energy of an electronic system expressed as an approximate density functional \( E[\rho] \) with respect to the electron density \( \rho \) or maximizing the relativistic proper time \( \tau \) of a classical particle with respect to a path through spacetime. Additionally, finding the points of stationary action that lead to the Euler-Lagrange equations of motion is often called "minimization of the action", but I can't recall ever having seen a proof that the action is truly minimized (as opposed to reaching a saddle point). This got me to think more about the conditions under which a functional is truly maximized or minimized as opposed to reaching a saddle point. Follow the jump to see more. I will frequently refer to concepts presented in a recent post (link here), including the relationships between functionals of vectors & functionals of functions. Additionally, for simplicity, all variables and functions will be real-valued.

Book Reviews: "Modern Liberty" by Charles Fried & "Development as Freedom" by Amartya Sen

I recently finished reading the books Modern Liberty by Charles Fried, and Development as Freedom by Amartya Sen. The first was part of the "Issues of Our Time" series, edited by Henry Louis Gates Jr., of which Whistling Vivaldi by Claude Steele (which I recently reviewed) is also a part; the second was one I chose as a comparison book mentioned in a review for the first on Amazon. Both deal with liberty and freedom, but in rather different ways. Below are my brief, decidedly nonspecialist thoughts on these books (content and style). After the jump will be some musings about issues related to these books. I am not a philosopher, nor am I an economist or sociologist, so many of the things that I say will probably be wrong or inconsistent; please feel free to point out issues in the comments below.

The book by Fried is a detailed defense of what appears (to my untrained eye) to be a centrist libertarian conception of liberty. The author essentially posits that human actions, and therefore the liberty to do so, are essentially individual in origin, so individual liberty should be held paramount, and can be taken to exist even in the absence of the state. He acknowledges the legitimacy of the state to tax and spend (even when the taxation is progressive), but argues that people should keep in mind that the state can simultaneously be the best friend and worst enemy of liberty. Additionally, he argues for equal application of the law as much as possible, and for the state to not interfere with people's choices when such interference would reduce choices that they would otherwise have no problem making, especially when those choices do not directly harm others in a criminal manner, focusing on three examples in particular throughout the book. With this in mind, he argues that the state should discourage certain behaviors through taxation rather than through heavy-handed direct intervention, and that stability in the laws and tax systems enforced is the least bad thing that the state can do with respect to liberty. Though I've focused specifically in pointing out things related to the state, he focuses on more abstract philosophical notions of liberty while simultaneously bringing the reader's attention to more mundane matters such as work, market transactions, and even sexual intercourse to better illustrate these points.

The book itself is fairly short, and moves along reasonably quickly (though I admit that I did get lost in some of the finer philosophical points). When introducing each justification for its conception of liberty, it also introduces each common criticism of that justification, and tackles almost every criticism head-on, without fear. There are quite a few things that I don't agree with about the ideas in the book. The overarching difference that I have is that this conception of liberty focuses on its origin within individuals, whereas I see the manifestation of liberty as being more dependent on societal contexts. An example of this would be in his contention that things like language, music, and culture originate within individuals; I would posit that these things would be meaningless for a single human in vacuum without any human contact, and it is only contact (and the history of such contact) with other humans that gives these things meaning. This is what I see as further leading to the author's general neglect of the consequences of liberty, choosing only to talk about the origins and processes of liberty; while I can see that this is philosophically consistent with the axiomatic treatment of the individual origin of liberty (and this also seems to be consistent with the desire for a static, predictable state due to its focus only on the unchanging processes of liberty, though I wouldn't agree with that either), the biggest issue that I have is that the author brings up the problem of the homeless person who has liberty but cannot make use of it if all property in that person's area is privately owned (and would therefore lead to the homeless person being kicked off of that property), but dances around this issue without really addressing it in a satisfactory way. With all of that said, I did enjoy reading this book overall, as it got me to think about the fundamental origins and processes of liberty in a new way, because before that, I was really only thinking about its manifestations/consequences.

The book by Sen is a longer exposition into how economic and political freedoms have to go hand-in-hand if they are to both be meaningful, and how human development is part and parcel of both. The author goes into how while utilitarian consequential formulations and libertarian process formulations of liberty are both important, both must be taken together instead of taking one or the other for liberty to be meaningful in the context of economics or politics. He further discusses how measures of economic development based solely on income or GDP/GNP per capita are quite flawed, so more nuanced, granular metrics are required, based on how different people's "functions" and "capabilities" operate, are fulfilled, and can be altered. Ultimately, he demonstrates that free markets and economic liberty, in conjunction with institutional corrections for certain glaring inequalities in capabilities, would allow the greatest human development leading to the greatest freedom.

Reading this book made me realize that I had intuitions for many ideas that the author clearly put into words (so I guess for now my lay economic views have a lot in common with those of Amartya Sen); in particular, I was already thinking about how Fried's neglect of the consequences of liberty made his treatment somewhat incomplete, and how humans being social animals means that the circumstances of one's political and economic existence cannot be ignored when considering the meaning of liberty, even before picking up Sen's book. However, there are a couple of issues that I have with this book. One is that there are several times where he repeats a point overly much. I don't mean that he just just repeats few words over the course of the book: I mean that he sometimes repeats entire multi-paragraph passages for no good reason, so the book could be a lot more terse and concise than it is. The other is that he argues that loss of income can affect a person mentally and physiologically in more lasting ways than simply by loss of purchasing power, especially if that person is ill, disabled, or so on, so he argues that specific institutional safety nets (presumably like social security, food stamps, and so on) and not simply lump-sum transfers of money (or increases in income) are necessary for countering poverty and promoting human development. I'd argue instead that if he can say that loss of income (and not just low income) is enough to push people into that downward spiral, then it would stand to reason that providing a strong enough guaranteed safety net through a simple money transfer (essentially, a minimum income, which I will discuss farther below) should be sufficient to prevent that, while simultaneously giving people the choice as independent agents to spend it as they like. Finally, at vary points, he promises to discuss how lessons from development in underdeveloped countries can be applied to the development of marginalized groups in more developed countries, yet as far as I could tell, that promise was never satisfactorily fulfilled. (Also, as a minor quibble: he makes reference to Madhavacharya having catalogued various schools of "Hindu" thought, including Buddhism, Jainism, and various atheist schools of thought, while calling him a Vaishnavite. A simple Wikipedia search shows that this Madhavacharya was a follower of Advaita philosophy, and was a separate person who was born only in the last few years of the life of the Dvaita founder Madhvacharya. It seems odd that Sen, who seems familiar with the "atheistic schools of Hinduism" catalogued by the later Madhavacharya, would make this error, so perhaps this was an oversight by the editor, or maybe the distinction between the two only became clear with scholarship after the publication of this book.)

In the Amazon review that I read, it seemed like these two books would oppose each other, but I would instead posit that they complement each other nicely. That brings me to the end of the review. Follow the jump to see some further thoughts on minimum income and related things that have been bouncing around in my head of late.

Reflection: Is the National Honors Society still about Honor?

After talking with a few friends and fellow students, I think I have fully (in my mind) fleshed out an opinion on our school's chapter of the National Honors Society. This, however, will solely focus on the recent cookie dough sale.
[Note]: Some of the things in this post may strike you as too controversial or offensive, and for that, I sincerely apologize. Though I have many grievances against the organization, (a) its core intentions are good and (b) it's not worth leaving when I only have a few months of school left so I am staying in the organization.
Furthermore, I want to say to those NHS officers who may be reading this that this is in no way meant as an attack on your person or character. This is meant to just be my thoughts about the organization as a whole, and if you find this offensive, I sincerely apologize. [/Note]
Other people in this organization (who also don't like the organization's current modus operandi) have said that the current state of affairs has only been true for the last year or 2. Before, the NHS was much more dedicated to actual community service, though they did not participate in as many events overall. Fundraising was not a priority.
Now, members must sell (under the threat of revocation of membership or other penalty) 4 tubs of cookie dough at $15.00 each (for those NHS graduates reading this, they raised the price by $1/tub). This is true for the fall and spring cookie dough sales. The fall one is understandable as the proceeds go to helping villages in Kenya. The spring one, though: do we really need that much money to continue operating? Or is Ms. Cresham just wasting a lot of money on who-knows-what?
Furthermore, why must members sell 4 tubs of cookie dough? I think it's perfectly fair to ask 15 hours of service each year towards NHS. This is the kind of service the organization should focus on. By contrast, selling requires people to want to buy cookie dough; this also requires finding buyers.
I know this may seem a little odd or offensive, but I just can't help but notice a cultural difference in selling cookie dough. Almost all of the "big sellers" that Ms. Cresham touts are Caucasian; props to them for their great sales. Yet, almost all of the people who complain about not being able to sell cookie dough are not Caucasian. From my own experience, cookie dough is not a hot seller with Indian families; yes, families might buy a tub, but it will last a very long time. By contrast, families here will buy a lot because they can finish that much by the time the next cookie dough sale comes around. My emphasis is not on how much one can eat but on how much a Caucasian American family vs. an immigrant Asian family actually wants the food. That said, I'm not trying to disparage Indian people - quite the opposite: maybe the NHS should try selling dough used for making samosas. Even then, though, there's a lot more emphasis (as far as I've seen) on cooking fresh food from scratch in Indian families (and presumably other Asian families as well), whereas here, families are a lot more receptive to premade foods like cookie dough. Even my cousin who bakes often makes her own dough; she doesn't use stuff like premade cookie dough.
Another thing to point out, as Ms. Cresham herself points out at the beginning of this cookie dough sale, is that the people who sell crazy amounts of dough do so because their parents have huge connections. This is perfectly fine with me; that said, not everyone is so well connected to people who would be so willing to fund an enterprise like the NHS.
Because of the issues of ethnicity (and the resulting affinity or revulsion towards premade cookie dough) and business connections (and the resulting success or failure to sell large amounts of cookie dough), why must students sell 4 tubs of dough? Many of the people who I talked to who were complaining about this minimum genuinely could not sell 4 tubs - they had to buy some (or all) themselves, and then they have to finish those tubs (without being able to cook them as no one in the house knows how); I was also in the position of having to buy a tub of cookie dough for myself (though the stuff is actually good, so this reason is only partially true to the argument).
I think it's fine that the organization is doing the cookie dough sale; I just think forcing members to sell a minimum number of tubs is too fraught with problems. I think the NHS should mandate members to participate in other activities for hours. I'm fine with them requiring members to do the Miniwalk (though I'm still not OK with the mandatory $15 for the (invariably oversized) shirt and donation). More such actual service requirements would be much greater appreciated (at least from my end).