유튜브 채널 WIRED
Computer Scientist Explains Machine Learning in 5 Levels of Difficulty | WIRED
Hilary Mason
- Computer Scientist
- Cofounder & CEO of Hidden Door
Machine Learning
- when we teach computers to learn patterns from looking at examples in data,
- such that they can recognize those patterns
- and apply them to new things that they haven't seen before.
5단계의 대화상대에게(어린아이~전문가) ML 설명하기
LEVEL 1. Child
ML
- a way that we teach computers to learn things about the world
- by looking at patterns and looking at examples of things.
개, 고양이, 늑대, 자칼, 사람 사진 보여주고
→ "Is it a cat or dog?'
→ 개, 고양이에서 벗어난 답을 했을 때 이유 묻기
ML
- When we teach machines to make guesses about what things are
- based on looking at a lot of different examples.
tests in school
- practice problems before the test
→ In the test, you're not seeing any problems that you don't know how to solve as long as you did all your practice.
→ Machines work the same way.
동물 사진을 천만 개 주고 개 or 고양이인지 구분하라 하면 어떻게 할 거니? 빨리 할 수 있겠어? → No
LEVEL 2. Teen
ML
- (Teen) Humans being able to teach machines or robots how to learn themselves.
- (Mason) When we teach machines to learn from data,
- to build a model from that data or a representation of that,
- and then to make a prediction.
Spotify - recommendation system
- "if you like Melanie Martinez, one of the other songs you might like is by Au/Ra."
What machines can understand
- The machine can understand whatever we tell it to understand.
- Things like the pitch or the pacing or the tone.
- Sometimes machines can figure out things about music or images or videos that we don't tell it to discover.
Facebook or Instagram use ML to target ads
- They know so much data. (Where you live, Where your device is)
- People in aggregate are actually pretty predictable.
- They're able to take that data that we already give them
- and make predictions based on that
- as to what ads they should show us.
Algorithms
- (Teen) A set of steps or a process carried out to complete something.
There are thing that machines are really great at that humans are actually not great at.
- watching every video posted on TikTok every day.
- A machine can analyze all of them and then make recommendations to us.
People are really great with only one or two examples of learning something new
- and incorporating that into our model of the world
- to make good decisions.
- Whereas machines often need tens of thousands of examples,
Machines are great at predicting based on what they've seen in the past,
- but they're not creative.
- They're not going to invent.
LEVEL 3. College Student
An undergraduate who study Math and Computer Science in New York University
Gmail program
- (Undergraduate) There would be a lot of machine learning models happening at once.
- (Mason) Models that are operating to do things like figure out if a new email is spam or not.
- (Mason) What would you think about if you were looking at an email and trying to decide if it went in one category or another?
- (Undergraduate) I'd probably look at certain keywords. Maybe if the recipient and the sender had exchanged emails before and generally, those fell into in the past.
- (Mason) So these are things we would call features. And we go through a process where we do feature engineering, where somebody looks at the example and says,
- "Okay, these are the things that I think might allow us to statistically tell the difference from something in one category versus another."
- (Mason) So for example, perhaps you don't speak Russian, you start getting a lot of email in Russian.
Supervised Learning classic classification approach
- A person would need to think about those features and creatively come up with them
- in approach we call the kitchen sink approach.
- which is just try everything you can possibly think of and see what works.
Unsupervised Learning
- We don't have labeled data and we're trying to infer some structure out of the data
- is you're projecting that data into a space and looking for things like clusters.
- And there's a bunch of really fun math about how you do that, how you think about distance
- and by distance, I mean that if we have two data points in space, how do we decide if they're similar or not?
How do the algorithms themselves usually differ between unsupervised and supervised learning?
- Supervised Learning, we have our labels and we're trying to figure out what statistically indicates
- if something matches one label or another label.
- Unsupervised Learning, we don't necessarily have those labels.
- That's the thing we're trying to discover.
Reinforcement Learning
- You can think about it like a turn in a game
- and you can play, you know, millions and millions trials
- so that you're able to develop a system
- that by experimenting with reinforcement learning
- can eventually learn to play these games pretty successfully.
- It also thrives in environments where you have a decision point, a pallette of actions to choose from.
- It actually comes historically from trying to train a robot to navigate a room.
- If it bonks into this chair, it can't go forward anymore.
- If it keeps exploring, it'll eventually get to the goal.
Deep Learning
- which is essentially using neural networks
- and very large amounts of data to eventually iterate on a network structure that can make predictions.
Is there a situation which you'd want to use a deep learning algorithm over a reinforcement learning algorithm?
- So typically, you would choose deep learning if you have sufficient high quality data, hopefully labeled in a useful way.
- If you really are happy not to necessarily understand or be able to interpret what your system is doing
- or you're willing to invest in another set of work afterwards to understand what the system is doing once you've already trained it.
- And this also comes down to the fact that some things are actually really easy to solve with linear regression or simple statistical approaches.
You could build a system that could actually be useless.
- A big telecom company → a data scientist built a deep learning system to predict customer churn. → It was very accurate, but it wasn't useful because nobody knew why the prediction was what it was.
- This is a good example of a very real world kind of machine Learning problems where the solutions to this was to build an interpretable system on top of the accurate predictions not to throw it away.
- but to do a bunch more work to figure out the why.
LEVEL 4. Grad Student
Graduate Student who is in her first year of a PhD in Computer Science and studying natural language processing and machine learning in Columbia University
What have you been working on or interested in lately?
- (graduate) I've been looking at understanding persuasion in online text and the ways that we might be able to automatically detect the intent behind that persuasion or who it's targeted at and what makes effective persuasive techniques.
What are some of the techniques you're applying to look at that debate data?
- (graduate) Something I'm interested in exploring is how well it works to use deep learning and sort of automatically extracted features from this text versus using some of the more traditional techniques that we have, things like lexicons or some sort of template matching techniques for extracting features from texts.
In the last few years, we've seen a lot of changes and improvements in the capabilities of NLP systems. So is there anything in that you're particularly excited about exploring further?
- I'm really interested in, sort of, the creative potential that we've started to see from NLP systems with things like GPT-3 and other really powerful language models.
- It's really easy to write long grammatical passages thinking about the way that we can then harness, like,
- the human ability to actually give meaning to those words and, sort of, provide structure
- and how we can combine those things with the, kind of like, generative capabilities of those models now is really interesting.
LEVEL 5. Expert
Claudia Perlich
- Computer Scientist
(Claudia) What types of biases in the data collection, and then also in usage?
- We now call it the bias, but we're still struggling with the society not really living up to its expectations
- and then machine learning bringing it to the forefront.
(Mason) When you're collecting data from the real world and then building machine learning systems that automate decisions based on that data,
- all of the biases and problems that are already in the real world then can be magnified through that machine learning system.
(Mason) And so, it's not just the provenance of that data, but it's, sort of, deeply understanding, "Why does it look the way it looks? Why was it collected this way? What are the limitations of it?"
(Mason) So things like actuarial science, operations research, where they actually are not using machine learning as much as you might think.
(Claudia) I am somewhat frustrated with a generation of students who have standard data sets that they never think about what the model needs to be used for.
- that they never have to think about how the data was collected.
