An Engineer's Literary Notebook

So spring comes again.  The unfurling daffodils are a welcoming sight after the long winter. And if you happen to wander lonely as a cloud, would the memory of daffodils past make you smile, or would you weep for the fading of the daffodils yet to come as Robert Herrick did a few centuries ago?  It could also happen that neither Wordworth’s nor Robert Herrick’s poem satisfies  and you want to train an artificial neural network to select a poem for you, a poem that matches what you long for at this very moment, by feeding it with copious quantity of  poems you ever plainly love.  Suppose that you train it with poems by Edna Saint Vincent Milay, Gerard Manley Hopkins, he of the dappled things,  Heather McHugh, Emily Dickson of the slant light, Huy Can, A. E. Stallings, Dylan Thomas along with the feeling that each poem invokes, creating the necessary  weight for  each feeling to quantify the bias for the neurons in the hidden layer(s) of the network. Of course this is an oversimplication of why we read poetry, but one must start somewhere.

Using feeling as an input parameter is fraught with subjectivity, but we are not traversing new territory. Pain has been graded on a scale of 1 to 10 so why couldn’t pleasure be measured? Pleasure is one of the reasons for poetry to endure.  Even if subjectivity is expected — the neural network we design here is selecting a poem especially for you — to quantify the pleasure a poem provides to create bias to the hidden layer of  our neural network is still difficult.  Which of these poems should have a larger bias?

Rhyme may be a parameter for some readers but not for others.  For this reader, it is.  I remember the first time being drawn to the poem that starts with these lines, and the  welcome attraction has stayed with me ever since. The pleasure it provides surprises me anew every time I read it.

Of course words matter.   Our artificial neural network learns the words in the poems we love to introduce to us poems that we will love.  Given the infinite number of nuances in the permutation of words, does this neural network need to be a deep neural network that has at least two hidden layers to be able to provide meaningful output of a higher quality than the “you may also like” recommendation of certain on-line merchant? 

The answer is uncertain because training deep neural networks is still largely done by trial and error. In fact, there have been talks of  machine learning as alchemy.

That machine learning is equated to alchemy is poetry in itself.  And it is fitting that we use a poetic device to select poetry for our enjoyment.  If we were to be successful in devising such a network, I hope that it will point me to this poem.

Thank you for the inspiration, dear muse.

Acknowledgements

1. The daffodils photo is from https://blog.thechanler.com/cliff-notes/2017/4/daffodils-for-days
2. The neural network diagram is from http://iwaponline.com/content/74/10/2497.figures-only
3. The trial and error nature of neural network is from https://stats.stackexchange.com/questions/182734/what-is-the-difference-between-a-neural-network-and-a-deep-neural-network-and-w
4. The machine learning as alchemy talk is at https://www.youtube.com/watch?v=Qi1Yry33TQE.
5. Many thanks to A.E Stallings and Matthea Harvey for writing my favorite poems.

Source: Of Beauty and Truth, Denied

Should an engineer attempt to prove the assertion that beauty is truth, truth beauty? I say yes even after realizing that the term negative capability also comes from the same source, the poet John…

Source: Of Beauty and Truth, Denied

Should an engineer attempt to prove the assertion that beauty is truth, truth beauty? I say yes even after realizing that the term negative capability also comes from the same source, the poet John…

Source: Of Beauty and Truth, Denied

If you were able to touch the wings of a dragonfly in flight one summer afternoon, would you be able to replicate the sensation of flying? Icarus not withstanding, intentionally directional weightl…

Source: Surprising The Senses

Last night I saw my father smiling broadly behind the wheel of a shiny Vetta. The chrome trims, the cream leather,  accents the color of ocean made my admiration for the automobile felt normal, Vetta a normally coveted make of cars believable the way memory recalled in dreams reinvented itself effortlessly, imposing itself nonchalantly. Where those fragments of memory go when the dreams recede I do not know. I only know that bits of that dream hung on as I woke. The familiar sense of loss returned, but this time a little less pronounced because I was busy reaching toward the source of those fragments, the hippocampus of dreams where my father still writes poetry.

Is it the hippocampus of dreams that I should seek? To answer the question I would need to know whether my quest is for a long term explicit or implicit memory.  The biology students among us already know that explicit memory involves conscious awareness, recording facts, events,  objects, people and places via the hippocampus and adjacent cortex, whereas implicit memory is acquired unconsciously to store perceptual and motor skills, requiring not the hippocampus but the cerebellum, the striatum, and the amygdala.   Do you find it difficult to reconcile the association of those tangible biological parts to such gossamer things like memory?  It has been shown experimentally that the human brain contains about 86 billions neurons.  A single neuron can have up to a thousand synapses, the units of information storage for short-term memory. The engineer in me dispassionately appraised the experiment’s method of counting neurons whereas that other part of me recited from memory the poems I read to my father, over and over again because his short term memory was not what it had been.

As far back as I could remember, my father wrote poetry.  He wrote each of us a poem to celebrate our births. He wrote about ordinary happenings such as the time when he showed his little sister the newly hatched chicks, about his empathy for the Quynh flowers that bloomed at midnight with no one watching, about running out of tea when a friend visited.  Later,  I think he found poetry cathartic as he tried to work out the helplessness he felt after finding refuge in another country, as evident in this poem.

The moon was a recurrent image in his poetry. Even toward the end when his memory had all but gone, he smiled when I read one of his favorite stanzas of “Chinh Phu Ngam” to him, the one where the drumbeats on the long rampart shook the moon.  By then he had stopped writing, and suddenly his numerous poems were no longer enough, sad poems, happy poems, and the poems he wrote for my mother, the love of his life for sixty years. She went first and I could not bear it when he asked after her, her for whom he wrote this poem:

I got up early and went to work in the garden this morning. The blue sky, clear and crisp, made it impossible to brood. The autumn joy sedium, confined to a container, had withered. But at its base I spied some new green. I lifted it up gently and placed it securely under a piece of earth across from the heritage rose. It will bloom again next year.  I thought about the many roses in my father’s garden. I still don’t know if there is an afterlife, but I am thankful that my father’s love of gardening and, especially, his love of poetry live on in me.

Thank you for helping me heal, dear Muse.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Acknowledgement:

1. The limpic drawing is from https://writersforensicsblog.wordpress.com/2013/05/17/ptsd-blame-your-hippocampusamygdala-complex.
2. The moon drawing is from http://www.acclaimclipart.com.
3. The sedium photograph is from http://www.edenbrothers.com.
4. Most of the brain and memory information is from http://www.cell.com/cell/abstract/S0092-8674(14)00290-6?_returnURL=http%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS0092867414002906%3Fshowall%3Dtrue.
5. The number of neurons is from http://www.nature.com/scitable/blog/brain-metrics/are_there_really_as_many.
6. The turquoise car photo is from https://www.pinterest.com/hrussom800208/mustang,