Time Travel: 2014

Chapter 203 Academic tool man get√

In this probabilistic latent semantic analysis model, polysemy words are grouped under different topics, while synonyms are grouped under the same topic.

This can avoid the impact of synonyms and polysemy on text similarity calculation.

However, the parameters of probabilistic latent semantic analysis models grow linearly with the number of documents.

It is easy to overfit and poor generalization.

This situation is largely due to the dimension explosion.

Because overfitting only occurs in two situations: predicting relatively few parameters in high-dimensional space and predicting many parameters in low-dimensional space.

A model proposed to avoid dimensionality explosion has dimensionality explosion.

It's somewhat tragic.

In fact, the above two models are not the only ones proposed.

Since then, many research teams have made varying degrees of attempts at the model level.

However, these models are either completely different from each other and are not conducive to dimensionality reduction.

Either it brings new problems while reducing dimensionality.

In short, these models all have various failings.

Even though Eve Carly believed Lin Hui was a genius.

But Eve Kali feels that it will be difficult for Lin Hui to find a new low-dimensional model that can avoid the dimension explosion despite countless previous attempts.

Eve Carly thought about it for a long time but still couldn't figure out what Lin Hui meant.

Eve Carly explained her thought process to Lin Hui.

Lin Hui listened very carefully.

After listening, Lin Hui smiled and said: "You have thought of so many situations of transformation from high dimensions to low dimensions.

And you also mentioned before that when machines recognize text, natural languages ​​are often digitized in order for machines to recognize natural languages.

Then, in order to distinguish the attributes of these values, they are further vectorized.

Now that you can understand this, you should know that problems involving natural language processing are prone to dimensionality explosion largely because the original data has extremely high dimensions.

In this case, why don’t we solve the problem directly from the source and do something about the original data? "

Hearing Lin Hui's words, Eve Carly seemed to be touched deep in her soul.

Eve Carly trembled: "You mean to directly perform low-dimensional processing on the original high-dimensional data?

After processing the low-dimensional data, perform semantic text similarity analysis on the architectural model? "

It saves time and energy to talk to smart people.

This is pretty much what Lin Hui wanted to express.

Nowadays, the most commonly used encoding method in this space and time is 1hot encoding.

Although this kind of coding has played a positive role for a long historical period.

But this encoding often brings a lot of trouble.

In the past few years, word vectors were basically encoded through distributed vectors.

Distributed encoding is equivalent to projecting the original data into a lower-dimensional space.

In this way, the original data is compressed and embedded from the sparse high-dimensional space into a lower-dimensional vector space.

This will undoubtedly be helpful for subsequent processing.

Of course, projection does not just mean projection.

It also needs to be trained by neural network learning.

As for how to train?

This is a purely technical matter.

In the past, the distributed representation of word vectors also had a proper name called word embedding.

Although what Lin Hui is talking to Eve Carly already involves some of the principles of distributed word vectors.

But Lin Hui is not afraid of Eve Carly plagiarizing ideas.

After all, what he said was only part of the principle.

Selective truths are sometimes more misleading than lies.

As for how much information Eve Carly can get from Lin Hui, it mainly depends on how well the two get along.

In fact, Lin Hui hopes to tell Eve Carly everything one day.

Because there is only one situation if Eve Carly can get all the information.

That is why she is willing to be Lin Hui's tool.

There is no need to hide anything from your tool.

Lin Hui also urgently needs a purely academic tool person.

After all, it would be ideal if someone in scientific research could provide guidance on how to get the paper.

If you go too deep into it, you may not become stronger, but you will definitely become bald.

The Eve Carly in front of her is smarter and more understanding.

Eve Carly is young and passionate, which is the most important quality that a good prospective scientific researcher should possess.

The most important thing is that Eve Carly seems to be very innocent and easy to deceive.

Such a simple sister.

It seems inappropriate not to trick her into writing a thesis...

It doesn't seem to be that troublesome.

Things went smoother than Lin Hui imagined.

Had an in-depth chat with Eve Carly.

(ps:... I have omitted a lot of words. Please add them slowly when you have time in the future. The added words are not counted in the billed words)

Eve Carly seemed determined.

He mustered up the courage to say to Lin Hui: "If, I mean, if possible, can I be your assistant?"

After hearing her words, Lin Hui seemed to be hesitant.

Eve Carly also knew that her request was a little offensive.

Although Eve Carly came up with this idea a little hastily.

But Eve Carly believed in the choices she made in her heart.

The conversation with Lin Hui gradually deepened, and Eve Carly seemed to be exposed to a brand new world for a while.

On the way here, Eve Carly felt Lin Hui's silent call to him: "Are you eager to open that door?"

After communicating with Lin Hui this time, after hearing what Lin Hui said just now, Eve Carly seemed to have witnessed Lin Hui push open the door.

Naturally, Eve Carly would not let go of the opportunity to open that door with Lin Hui.

Seeing that Lin Hui seemed hesitant, Eve Kali hurriedly said: "I am willing to sign an agreement, and I will keep your research results strictly confidential..."

Eve Carly continued: "I don't need scientific research funds either..."

A scholar who traveled thousands of miles across the ocean had such a respectful attitude.

Lin Hui would seem very unkind if he didn't agree.

Lin Hui decisively agreed to Eve Carly's request.

The agreement must be signed, Lin Hui didn't want the results to be stolen.

However, scientific research funding and corresponding remuneration must still be paid.

Employees can use their love to generate electricity.

But the boss can't really let employees use love to generate electricity.

It would be easy to die suddenly on the spot.

Both Lin Hui and Eve Carly are willing to conclude an academic cooperation agreement.

In this case, things went smoothly.

Lin Hui and Eve Carly made an agreement easily.

Of course, the reason why the academic cooperation agreement could be concluded so smoothly was because both Lin Hui and Eve Carly had the intention.

It is also due to the fact that there is a special set of standards for academic cooperation in this world of time and space.

In this time and space, countries around the world still continue traditional practices except for academic cooperation involving confidential projects.