Journal

On Thursday, we met with Dr. Hassibi to talk about what we have been doing after last time. What I did before our meeting was trying to tinker with the eigenvalues and eigenvectors. In my attempt to do Kmeans on the dataset, I stored the first k-number of eigenvectors and vertically stacked them on to each other into another column matrix. I then ran Kmeans on it but that was not the correct execution as pointed out by Dr. Hassibi. He said to take the eigenvectors and put them next to each other (stacking horizontally) and then run it. It makes sense since stacking vertically would mean there would be k*n # of nodes. Here is the photo when we plotted our results and below is a video of our meeting.

A graph showing Kmeans clustering nodes into three classes.

As for the future, we were given two options: keep experimenting with the dataset at hand or try out a new one. With the current data, much is there to be explored such as test the question of whether a pair-wise or trio-wise comparison is the best method of "classification" if you will. I also want to try out a semi-crowdsourced approach but I don't think it's viable because there doesn't exist a dataset with the same species and format. For new datasets, we proposed the idea of making our own dataset and asking some thirty volunteers to make comparisons. After our discussion on Friday, I decided that I would have a minimal role in organizing it and more on trying things with the dataset.

Last week, April 11th to 12th, Ben, Jeffrey, Matthew, and I finally hosted our online highschool hackathon, TitanHacks. We had around 387 signups (a lot from outside of the U.S.), 171 registrants, 135 of them submitted, and 50 projects in total. Setting up the entire thing and running it was such a unique experience and I think we will be doing it next year for sure. However, it will be in-person and LA because online was such a pain. There were several unethical hackers and cheaters that we caught. Also, shipping costs were gonna hurt our budget until we switched to eGift cards. Below is a winning submission video:


In other news, I have found my new favorite drink, Milkis. It's a carbonated drink that takes the flavor of milk and yogurt and maybe a fruit. I found it in HMart while shopping for "essentials" and bought a box of them. From it, there was regular-flavored, melon-flavored, apple-flavored, and strawberry-flavored, my favorite. Despite the beverage having an unusual vibrant color (like with the strawberry one, it's a pink-red), I still like it. One thing I criticize it for is that it's super artificial. No juice and it has a very exaggerated flavor of the fruit. If you drink it once a day at max, you'll be fine.

Picture of the drink.

Comments

Post a Comment

Popular posts from this blog

Technical Journal - Research Papers

Image
On the Efficiency of Data Collection for Crowdsourced Classification :        The paper aims to explain the accuracy gaps between popular collection policies (state-action mapping) on crowdsourced data. The quality of crowdsourced data is often highly variable so it is suspected that it is the cause. However, studies show that the policies used to collect such data have a strong impact on the accuracy of the system. The first theoretical explanation of the accuracy gaps is that the non-adaptive uniform allocation, and the adaptive uncertainty sampling and information gain maximization. Done with the representation of the collection process in terms of random walks in the log-odds (log of odds of guessing correctly/incorrectly) domain, the author derives lower and upper bounds on the accuracy of the policies. The bounds accurately tell us by how much do two adaptive policies trump a non-adaptive one. It is believed that techniques used by them are applicable to addi...
Read more

Journal

Image
On Tuesday, March 24th, we had a meeting with Professor Hassibi over Zoom. Prior to our talk, we had issues with our code regarding the fact that it didn't produce the results expected. He told us to find and graph the eigenvalues of the graph Laplacian, as that would indicate the number of clusters found and value you can refer other datapoints to. He noted that graphing the corresponding eigenvector would indicate whether a cluster had been isolated or not. On that note, it was up to us to tinker with values for a bit and try different algorithms like K-means. Below is a video of the meeting with Professor Hassibi: This week has been very interesting, to say the least. As you may know, COVID-19 has just taken the U.S. for a turn and has put a pause on daily life, forcing businesses to adapt, overcome, or fail in some cases. For us high schoolers, classes have migrated to the Zoom platform. With it followed teachers with bad internet service, leading to indecipherable dialect...
Read more

First Month

Image
The first four weeks of this Caltech Stem class have been interesting, to say the least. Much of the material we went into have been all over the place with various objectives. We started off with teamwork exercises, watching college students present their research, replicating one for ourselves, writing a concept map, and going to Caltech to meet Dr. Hassibi. We had two, teamwork workshops for the first two weeks. One of which was the dominoes problem where each student was given two dominoes and we were instructed to get three dominoes per person by exchanging dominoes for chips and chips for dominoes. Instructions were very vague other than that everyone must have a domino to "survive" a round. We struggled in the first few attempts but through several days of discussion and trial & error, we were able to find a way to accomplish our goal. Some key underlying goals of this exercise were the importance of leadership, communication, and open-mindedness...
Read more