BuboFlash - helps with learning

Edited, memorised or added to reading queue

Do you want BuboFlash to help you learning these things? Click here to log in or create user.

Flashcard 7629450120460

Tags

#English #vocabulary

Question

[...]

adjective

involving or requiring strenuous effort; difficult and tiring

Answer

arduous

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
arduous /ˈɑːdjʊəs,ˈɑːdʒʊəs/ Learn to pronounce adjective involving or requiring strenuous effort; difficult and tiring

Original toplevel document

Open it
arduous /ˈɑːdjʊəs,ˈɑːdʒʊəs/ Learn to pronounce adjective involving or requiring strenuous effort; difficult and tiring "an arduous journey" Similar: onerous taxing difficult hard heavy laborious burdensome

Flashcard 7629883182348

Tags

#causality #statistics

Question

If all the paths between two nodes 𝑋 and 𝑌 are [...], then we say that 𝑋 and 𝑌 are d-separated.

Answer

blocked

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
If all the paths between two nodes 𝑋 and 𝑌 are blocked, then we say that 𝑋 and 𝑌 are d-separated.

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7629890522380

Tags

#causality #statistics

Question

A potential outcome 𝑌(𝑡) is distinct from the [...] outcome 𝑌 in that not all potential outcomes are observed

Answer

observed

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
A potential outcome 𝑌(𝑡) is distinct from the observed outcome 𝑌 in that not all potential outcomes are observed

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669030718732

Tags

#deep-learning #keras #lstm #python #sequence

Question

The goal of the backpropagation training algorithm is to modify the [...] of a neural network in order to minimize the error of the network outputs compared to some expected output in response to corresponding inputs.

Answer

weights

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
The goal of the backpropagation training algorithm is to modify the weights of a neural network in order to minimize the error of the network outputs compared to some expected output in response to corresponding inputs.

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669035699468

Tags

#feature-engineering #lstm #recurrent-neural-networks #rnn

Question

Third, while LSTM models offer a markedly improved solution to the problem of exploding gradients (over vanilla RNN models), they are not guaranteed to be shielded from it entirely. Facing such an issue, the analyst might need to rely on computational tricks, such as [...] (Bengio, 2012), gradient scaling, or batch normalization (Bjorck, Gomes, Selman, & Weinberger, 2018)

Answer

gradient clipping

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
of exploding gradients (over vanilla RNN models), they are not guaranteed to be shielded from it entirely. Facing such an issue, the analyst might need to rely on computational tricks, such as <span>gradient clipping (Bengio, 2012), gradient scaling, or batch normalization (Bjorck, Gomes, Selman, & Weinberger, 2018) <span>

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669038320908

Tags

#deep-learning #keras #lstm #python #sequence

Question

The additional hidden layers are understood to recombine the learned representation from prior layers and create new representations at high levels of [...].

Answer

abstraction

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
The additional hidden layers are understood to recombine the learned representation from prior layers and create new representations at high levels of abstraction.

Original toplevel document (pdf)

cannot see any pdfs

Annotation 7669051690252

#deep-learning #keras #lstm #python #sequence

1.4.1 LSTM Weights

Output Weights. Used to weight the output from the last time step.

status	not read	reprioritisations
last reprioritisation on		suggested re-reading day
started reading on		finished reading on

Parent (intermediate) annotation

Open it
l has weight parameters for the input, output, as well as an internal state that is built up through exposure to input time steps. Input Weights. Used to weight input for the current time step. <span>Output Weights. Used to weight the output from the last time step. Internal State. Internal state used in the calculation of the output for this time step <span>

Original toplevel document (pdf)

cannot see any pdfs

Annotation 7669054573836

#deep-learning #keras #lstm #python #sequence

There are some practical considerations when scaling sequence data.

Estimate Coefficients
You can estimate coefficients (min and max values for normalization or mean and standard deviation for standardization) from the training data. Inspect these first-cut estimates and use domain knowledge or domain experts to help improve these estimates so that they will be usefully correct on all data in the future.

status	not read	reprioritisations
last reprioritisation on		suggested re-reading day
started reading on		finished reading on

Parent (intermediate) annotation

Open it
3.1.3 Practical Considerations When Scaling There are some practical considerations when scaling sequence data. Estimate Coefficients You can estimate coefficients (min and max values for normalization or mean and standard deviation for standardization) from the training data. Inspect these first-cut estimates and use domain knowledge or domain experts to help improve these estimates so that they will be usefully correct on all data in the future. Save Coefficients You will need to scale new data in the future in exactly the same way as the data used to train your model. Save the coefficients used to file and load them later when

Original toplevel document (pdf)

cannot see any pdfs

Annotation 7669057457420

#deep-learning #keras #lstm #python #sequence

LSTM’s special strength, namely, to learn to remember single events for very long, unknown time periods

status	not read	reprioritisations
last reprioritisation on		suggested re-reading day
started reading on		finished reading on

Parent (intermediate) annotation

Open it
A time window based MLP outperformed the LSTM pure-[autoregression] approach on certain time series prediction benchmarks solvable by looking at a few recent inputs only. Thus LSTM’s special strength, namely, to learn to remember single events for very long, unknown time periods, was not necessary

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669059554572

Tags

#recurrent-neural-networks #rnn

Question

The challenge for deep learning models of customer behavior remains their [...] nature and the lack of simple ways to interpret their behavior, which is especially true for the complex temporal dynamics of RNNs.

Answer

opaque

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
The challenge for deep learning models of customer behavior remains their opaque nature and the lack of simple ways to interpret their behavior, which is especially true for the complex temporal dynamics of RNNs.

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669061389580

Tags

#recurrent-neural-networks #rnn

Question

Embedding layers are used to reduce data dimensionality, compressing large vectors of values into relatively smaller ones, to both reduce noise and limit the number of model [...] required

Answer

parameters

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
Embedding layers are used to reduce data dimensionality, compressing large vectors of values into relatively smaller ones, to both reduce noise and limit the number of model parameters required

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669063224588

Tags

#feature-engineering #lstm #recurrent-neural-networks #rnn

Question

When an analyst uses feature engineering to predict behavior, the performance of the model will depend greatly on the analyst's domain [...]

Answer

knowledge

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
When an analyst uses feature engineering to predict behavior, the performance of the model will depend greatly on the analyst's domain knowledge

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 7669065059596

Tags

#deep-learning #keras #lstm #python #sequence

Question

For example, we can define an LSTM hidden layer with [...] memory cells followed by a Dense output layer with 1 neuron as follows:

model = Sequential()
model.add(LSTM(2))
model.add(Dense(1))

Answer

status	not learned	measured difficulty	37% [default]	last interval [days]
repetition number in this series	0	memorised on		scheduled repetition
scheduled repetition interval		last repetition or drill

Parent (intermediate) annotation

Open it
For example, we can define an LSTM hidden layer with 2 memory cells followed by a Dense output layer with 1 neuron as follows: model = Sequential() model.add(LSTM(2)) model.add(Dense(1))

Original toplevel document (pdf)

cannot see any pdfs

Edited, memorised or added to reading queue

on 01-Dec-2024 (Sun)

Parent (intermediate) annotation

Original toplevel document

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)

Parent (intermediate) annotation

Original toplevel document (pdf)