For this project, I created an instruction manual to guide users on the basic use of Python 3.8.0.
Huzaifa Anas Expanded Definition of Algorithm
TO: Prof. Jason Ellis
FROM: Huzaifa Anas
SUBJECT: Expanded Definition of Algorithm
Algorithms play a pivotal role in computer science and this paper by examining the etymology, definition, context, a working definition will be formed through analysis of different sources like dictionaries, journals, newspaper articles, and encyclopedias, blogs, and social media. Multiple sources are used because they all cater to different audiences. We first focus on the modern definition and its etymology to provide a basis to understand the term and then look at how the term is used practically for a holistic understanding of the term and finally attempt to formulate a definition based on these two factors to make a definition which works for Computer Science (CS).
The word algorithm originates as a homage to the famous al-Khwarizmi, who was a famous polymath remembered for his pioneering Algebra. The word algorithm was algorism in Middle English, which traces its roots to French words algorithme “ meaning “Arabic system of computation,” originating (under mistaken connection with Greek arithmos “number”) from Old French algorisme meaning “the Arabic numeral system “(etymoline.com). The progenitor of these was algorismus which was an archaic Latin translation of al-Khwarizmi. Eventually, during the mid 20th century, the word expanded to any method of computation.
The Oxford English Dictionary (OED) defines algorithms for Mathematics and Computing as “A procedure or set of rules used in calculation and problem-solving; (in later use spec.) a precisely defined set of mathematical or logical operations for the performance of a particular task”, and this concept is basically translated over to psychological use of the term (algorithm n.2 & 3., 2019). While looking at the OED’s historical definitions we see the definition limited to mathematical theorems, but then expanding. Overall the idea remained mostly static of having a systematic method to tackle some issues. Britannica also similarly defines an algorithm as a “systematic procedure that produces—in a finite number of steps—the answer to a question or the solution of a problem” (algorithm, 2019). The Gale Encyclopedia of Science defines it slightly differently as “a set of instructions for accomplishing a task that can be couched in mathematical terms. If followed correctly, an algorithm guarantees successful completion of the task”(Algorithm, 2014). Overall in these definitions, we see the trend of having a specific modus operandi to solve specific problems. The only real difference is the scope of this concept being applied from like computer science (CS), CS & math, or this concept in general.
Algorithms are ubiquitous in computer science and have numerous different techniques that drive them. Some are rigid, and some serve one purpose, while others are multi-purpose and constantly improve on their own. Every year we have algorithms being used in new frontiers, like the recent developments in facial recognition and medicine. Nowadays an algorithm can apply to multiple issues like with “AlphaZero[applying to] games of chess and shogi, as well as Go, by using the same algorithm and network architecture for all three games.”, which is different from the rigid OED definition of limiting to a particular task (Silver et al., 1). This is through the introduction of more advanced Artificial Intelligence (AI) principals, which allows an algorithm to do multiple tasks using its initial guidelines. AI algorithms, in general, don’t follow the traditional definition because “the area of AI called machine learning, in which computers derive expertise from experience, the training data for an algorithm can influence its performance.”, so even if the steps are followed correctly faulty training data can cause failure (Huston, 1). Another diverging feature of modern algorithms is that sometimes they build themselves like Jeff06 who “was dissatisfied by how long my genetic algorithm took to evolve a basic control scheme” after he set up the skeletal framework, compared to the traditional algorithms which are precise with instructions (https://www.codingame.com/blog). The goal and definition an algorithm mandates proper execution of an algorithm will result in the issue always being properly addressed, but this can be debated because there is no “right to examine and challenge the data used to make algorithmic decisions about us”, which can put into question the results for non-hard science algorithms because of manipulation and about whether the process itself is correct to fit all contexts, and the way results are categorized or given (Augwin, 2016). We see a similar situation in machine translation where “Google Translate turns non-gendered Turkish sentences  sexist sentences in English[because the] Google Translate  algorithm [bases] its translations on observed frequency of usage”, where although the algorithm is properly functioning and should give correct results, in reality, we don’t get our desired results, because of biased data or inherent shortcomings of the multipurpose algorithm. (Shams, 2017). From the rise of AI, we see algorithms becoming more fluid and adaptable, but also less consistent which is a big divergence from old rigid algorithms which we typically associate them with. Overall in the real world without perfect information, these so-called perfect systems won’t always properly perform but even the results themselves can be debated sometimes. All in all, in these quotes we see some form of divergence from the traditional definition of algorithms from their evolving nature, inconsistent performance, contentious parameters or results, and greater flexibility of each algorithm we see a more capable painting of the term, but with one with shakier foundations than these established definitions.
Algorithms now have become dynamic nature, possibly inconsistent, adaptable, and more independent through the Ai in computer science they should be defined as a set of instructions or guidelines, that may self regulate, which can be used to attempt to solve specific or group of problems with a large degree of success.
A Genetic Algorithm to Pilot Pod Racers. (2018, August 16). Retrieved October 14, 2019, from https://www.codingame.com/blog/genetic-algorithms-coders-strike-back-game/.
algorithm (n.). (n.d.). Retrieved October 15, 2019, from https://www.etymonline.com/word/algorithm.
algorithm, n.2. (2019). In Oxford English Dictionary Online. Retrieved from www.oed.com/view/Entry/4959
algorithm, n.3. (2019). In Oxford English Dictionary Online. Retrieved from www.oed.com/view/Entry/4959
Algorithm. (2019). In Encyclopædia Britannica. Retrieved from https://www.britannica.com/science/algorithm.
Algorithm. (2014). In K. L. Lerner & B. W. Lerner (Eds.), The Gale Encyclopedia of Science (5th ed., Vol. 1, p. 131). Farmington Hills, MI: Gale. Retrieved from https://link-gale-com.citytech.ezproxy.cuny.edu/apps/doc/CX3727800076/GVRL?u=cuny_nytc&sid=GVRL&xid=7b6e67ce (This one won’t work without the citytech.ezproxy.cuny.edu part)
Angwin, J. (2016, August 1). Make Algorithms Accountable. Retrieved October 14, 2019, from https://www.nytimes.com/2016/08/01/opinion/make-algorithms-accountable.html?searchResultPosition=3.
Hutson, M. (2018). Artificial intelligence faces reproducibility crisis. Science (New York, N.Y.), 359(6377), 725-726. https://science.sciencemag.org/content/359/6377/725
Shams, A. (2017, November 27). Facebook post by Alex Shams on the gender bias of google translates algorithm for Turkish to English. Retrieved October 15, 2019, from https://www.facebook.com/alexrezashams/posts/10105798694396775.
Silver, D. davidsilver@google. co., Hubert, T., Schrittwieser, J., Antonoglou, I., Lai, M., Guez, A., … Hassabis, D. dhcontact@google. co. (2018). A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play. Science, 362(6419), 1140–1144. https://doi-org/10.1126/science.aar6404
TO: Prof. Jason W. Ellis
FROM: Huzaifa Anas
DATE: September 17
SUBJECT: 500-Word Summary of Hassabis et al “Neuroscience-Inspired Artificial Intelligence”
Hassabis et al in Neuron argues that the field of neuroscience and AI (artificial intelligence) have a symbiotic relationship, but it’s in jeopardy, because of decreasing communication and collaboration. The contention states neuroscience provides a productive source of inspiration for algorithms and architecture, which is “independent of and complementary to the mathematical and logic-based methods and ideas that have largely dominated traditional approaches to AI” and “neuroscience can provide validation of AI techniques that already exist.” (Hassabis et al, 2017, p. 1). Moreover, they believe the progress in AI will eventually pay dividends to neuroscience by being a good test field. Within this article, past breakthroughs are examined to support this argument, while looking at how continued collaboration and communication can benefit both fields.
Two of AI’s backbones originate from neuroscience, which’s deep learning and reinforcement learning. Deep learning has revolutionized AI through dramatic advances in its neural and capable networks of learning freely from unstructured or unlabeled data. Reinforcement learning, the second pillar of modern AI, is a powerful tool enabling AI researchers to create software agents that act in an environment maximizing some sort of reward. In the 1940s artificial neural networks were developed, which could compute logical functions and ultimately “learn incrementally via supervisory feedback (Rosenblatt, 1958) or efficiently encode environmental statistics in an unsupervised fashion” (Hasabis, 2017, p. 2). This is the foundation for deep learning. Soon after backpropagation algorithms were made, which allowed learning to occur in networks of multiple layers whose value was recognized in 1986 by cognitive and neuroscientists working on Parallel distributed processing or PDP, which better-represented human-like behavior than serial logical processing, which AI researchers were focusing on. PDP has been applied to machine translation through the idea that “words and sentences can be represented in a distributed fashion (i.e., as vectors)” (Hasabis, 2017, p. 2). Deep learning ultimately became a field independent of PDP. Reinforcement learning comes from animal learning research, which Pavlov and Skinner pioneered. Reinforcement learning is used in robotic control, skillful play in backgammon and go.
If someone looks closely, AI research is still heavily inspired and guided by Neuroscience through AI work on attention, while eventually pivoting towards efficient learning and more independent behavior like transfer learning and imagination. The goal of AI is to form human-like behavior, and it’s practical an accurate biological framework as a reference. Attention is a critical issue currently because not all information is equal and therefore unlike before where all information was treated equally in neuroscience now information is being given different values, which allows for more efficient computing power usage. For the future, we want to decrease the computing power and a large amount of data needed for AI as currently. Humans can learn from a few examples, which AI can’t, and researchers are trying to apply developmental psychology ideas here. For imagination and transfer, learning neuroscience is still pioneering this part, but in the future, it’ll hopefully provide practical insights for AI work. All things considered, both fields can provide feedback to each other by having neuroscience provide ideas, and AI proves as a testing ground for these ideas. This isn’t compulsory, but just an effective and logical symbiotic relationship.
Article Cited APA format
Hassabis, D., Kumaran, D., Summerfield, C., & Botvinick, M. (2017). Neuroscience-inspired artificial intelligence. Neuron, 95(2), 245-258.
I’m not sure if restructuring definitions is considered plagiarism.