![Page 1: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/1.jpg)
Dive Deeper in Finance
GTC 2017 – San José – California
Daniel EgloffDr. sc. math.Managing Director QuantAleaMay 7, 2017
![Page 2: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/2.jpg)
Today
▪ Generative models for financial time series
– Sequential latent Gaussian Variational Autoencoder
▪ Implementation in TensorFlow
– Recurrent variational inference using TF control flow operations
▪ Applications to FX data
– 1s to 10s OHLC aggregated data
– Event based models for tick data is work in progress
![Page 3: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/3.jpg)
Generative Models and GPUs
▪ What I cannot create, I do not understand (Richard Feynman)
▪ Generative models are recent innovation in Deep Learning
– GANs – Generative adversarial networks
– VAE – Variational autoencoders
▪ Training is computationally demanding
– Explorative modelling not possible without GPUs
![Page 4: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/4.jpg)
Deep Learning
▪ Deep Learning in finance is complementary to existing models and not a replacement
▪ Deep Learning benefits
– Richer functional relationship between explanatory and response variables
– Model complicated interactions
– Automatic feature discovery
– Capable to handle large amounts of data
– Standard training procedures with back propagation and SGD
– Frameworks and tooling
![Page 5: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/5.jpg)
Latent Variable – Encoding/Decoding
▪ Latent variable can be thought of a encoded representation of x
Encoder
𝑝 𝑧 𝑥𝑥
Decoder
𝑝 𝑥 𝑧𝑥𝑧
𝑝 𝑧
▪ Likelihood serves as decoder
▪ Posterior provides encoder
![Page 6: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/6.jpg)
Intractable Maximum Likelihood
▪ Maximum likelihood standard model fitting approach
𝑝 𝑥 = 𝑝 𝑥 𝑧 𝑝 𝑧 𝑑𝑧 → max
▪ Problem: marginal 𝑝 𝑥 and posterior
𝑝 𝑧 𝑥 =𝑝 𝑥 𝑧 𝑝 𝑧
𝑝 𝑥
are intractable and their calculation suffers from exponential complexity
▪ Solutions
– Markov Chain MC, Hamiltonian MC
– Approximation and variational inference
![Page 7: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/7.jpg)
Variational Autoencoders
▪ Assume latent space with prior 𝑝 𝑧
𝑥 𝑥𝑧
𝑝 𝑧
𝑝 𝑧 𝑥 𝑝 𝑥 𝑧
![Page 8: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/8.jpg)
Variational Autoencoders
▪ Parameterize likelihood 𝑝 𝑥 𝑧 with a deep neural network
𝑥 𝑥𝑧
𝑝 𝑧
𝜇
𝜎
𝑝𝜑 𝑥 𝑧
𝑝 𝑧 𝑥
![Page 9: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/9.jpg)
Variational Autoencoders
▪ Parameterize likelihood 𝑝 𝑥 𝑧 with a deep neural network
▪ Approximate intractable posterior 𝑝 𝑧 𝑥 with a deep neural network
𝑥 𝑥𝑧
𝑝 𝑧
𝜇
𝜎
𝑝𝜑 𝑥 𝑧
𝑝 𝑧 𝑥
𝑞𝜃 𝑧 𝑥
𝜇
𝜎
![Page 10: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/10.jpg)
Variational Autoencoders
▪ Parameterize likelihood 𝑝 𝑥 𝑧 with a deep neural network
▪ Approximate intractable posterior 𝑝 𝑧 𝑥 with a deep neural network
▪ Learn the parameters 𝜃 and 𝜑 with backpropagation
𝑥 𝑥𝑧
𝑞𝜃 𝑧 𝑥
𝜇
𝜎
𝜇
𝜎
𝑝𝜑 𝑥 𝑧
𝑝 𝑧
![Page 11: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/11.jpg)
▪ Which loss to optimize?
▪ Can we choose posterior from a flexible family of distributions Q byminimizing a distance to real posterior?
Variational Inference
𝑞∗ 𝑧 𝑥 = argmin𝜃∈𝑄
𝐾𝐿 𝑞𝜃 𝑧 𝑥 ฮ𝑝𝜑 𝑧 𝑥
𝐾𝐿 𝑞𝜃 𝑧 𝑥 ฮ𝑝𝜑 𝑧 𝑥 = 𝐸𝑞𝜃 𝑧 𝑥 log 𝑞𝜃 𝑧 𝑥 − 𝐸𝑞𝜃 𝑧 𝑥 log 𝑝𝜑 𝑥, 𝑧 +
log 𝑝𝜑 𝑥≥ 0
Can be made small if Q is flexible enough
▪ Problem: not computable because it involves marginal 𝑝𝜑 𝑥
![Page 12: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/12.jpg)
▪ Which loss to optimize?
▪ Can we choose posterior from a flexible family of distributions Q byminimizing a distance to real posterior?
Variational Inference
𝑞∗ 𝑧 𝑥 = argmin𝜃∈𝑄
𝐾𝐿 𝑞𝜃 𝑧 𝑥 ฮ𝑝𝜑 𝑧 𝑥
0 ≤ 𝐸𝑞𝜃 𝑧 𝑥 log 𝑞𝜃 𝑧 𝑥 − 𝐸𝑞𝜃 𝑧 𝑥 log 𝑝𝜑 𝑥, 𝑧 + log 𝑝𝜑 𝑥
−𝐸𝐿𝐵𝑂(𝜃, 𝜑)
▪ Drop left hand side because positive
![Page 13: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/13.jpg)
▪ Which loss to optimize?
▪ Can we choose posterior from a flexible family of distributions Q byminimizing a distance to real posterior?
Variational Inference
𝑞∗ 𝑧 𝑥 = argmin𝜃∈𝑄
𝐾𝐿 𝑞𝜃 𝑧 𝑥 ฮ𝑝𝜑 𝑧 𝑥
𝐸𝐿𝐵𝑂(𝜃, 𝜑) ≤ log 𝑝𝜑 𝑥
▪ Obtain tractable lower bound for marginal
▪ Training criterion: maximize evidence lower bound
![Page 14: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/14.jpg)
▪ To interpret lower bound, write it as
Variational Inference
= 𝐸𝑞𝜃(𝑧|𝑥) log 𝑝𝜑 𝑥 𝑧 − 𝐾𝐿 𝑞𝜃 𝑧 𝑥 ԡ𝑝 𝑧
Reconstruction score
𝑧~𝑞𝜃 𝑧 𝑥𝑥
𝑝𝜑 𝑥 𝑧
Penalty of deviation from prior
log 𝑝𝜑 𝑥 ≥ 𝐸𝐿𝑂𝐵 𝜃, 𝜑
▪ The smaller the tighter the lower bound𝐾𝐿 𝑞𝜃 𝑧 𝑥 ฮ𝑝𝜑 𝑧 𝑥
![Page 15: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/15.jpg)
Applications to Time Series
▪ Sequence structure for observable and latent factor
▪ Model setup
– Gaussian distributions with parameters calculated from deep recurrent neural network
– Prior standard Gaussian
– Model training with variational inference
![Page 16: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/16.jpg)
Inference and Training
𝜇𝑡𝜎𝑡
𝜇𝑡−1𝜎𝑡−1
ℎ𝑡+1
ℎ𝑡+1
𝑧𝑡+1
𝑥𝑡+1
𝜇𝑡+1𝜎𝑡+1
𝜇𝑡𝜎𝑡
𝑥𝑡−1
𝑧𝑡−1
ℎ𝑡−1
ℎ𝑡−1
𝜇𝑡−1𝜎𝑡−1
𝜇𝑡+1𝜎𝑡+1
𝑥𝑡
ℎ𝑡
ℎ𝑡
𝑧𝑡
𝑞𝜃 𝑧 𝑥
![Page 17: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/17.jpg)
▪ Probability distributions factorize
Implied Factorization
𝑝𝜑 𝑥≤𝑇 𝑧≤𝑇 =ෑ
𝑡=1
𝑇
𝑝𝜑 𝑥𝑡 𝑥<𝑡 , 𝑧≤𝑡 =ෑ
𝑡=1
𝑇
𝑁 𝑥𝑡 𝜇𝜑 𝑥<𝑡, 𝑧≤𝑡 , 𝜎𝜑 𝑥<𝑡, 𝑧≤𝑡
▪ Loss calculation
– Distributions can be easily simulated to calculate expectation term
– Kullback Leibler term can be calculated analytically
𝑞𝜃 𝑧≤𝑇 𝑥≤𝑇 =ෑ
𝑡=1
𝑇
𝑞𝜃 𝑧𝑡 𝑥<𝑡, 𝑧<𝑡 =ෑ
𝑡=1
𝑇
𝑁 𝑧𝑡 𝜇𝜃 𝑥<𝑡, 𝑧<𝑡 , 𝜎𝜃 𝑥<𝑡, 𝑧<𝑡
![Page 18: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/18.jpg)
▪ Loss calculation
– Kullback Leibler term can be calculated analytically
– For fixed 𝑡 the quantities 𝜇𝜑, 𝜇𝜃, 𝜎𝜑, 𝜎𝜃 depend on
𝑧𝑡~𝑁 𝑧𝑡 𝜇𝜃 𝑥<𝑡, 𝑧<𝑡 , 𝜎𝜃 𝑥<𝑡, 𝑧<𝑡
– Simulate from this distribution to estimate expectation with a sample mean
Calculating ELBO
𝐸𝐿𝐵𝑂 𝜃, 𝜑 = −𝐸𝑞 ቂ
ቃ
σ𝑡 ቄ
ቅ
𝑥𝑡 − 𝜇𝜑𝑇𝜎𝜑
−1 𝑥𝑡 − 𝜇𝜑 + logdet 𝜎𝜑 +
𝜇𝜃𝑇𝜇𝜃 + 𝑡𝑟𝜎𝜃 − log det 𝜎𝜃
Approximate with Monte Carlo sampling from 𝑞𝜃 𝑧≤𝑇 𝑥≤𝑇
![Page 19: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/19.jpg)
Generation
𝜇𝑡𝜎𝑡
ℎ𝑡+1
𝑧𝑡+1
𝑥𝑡+1
𝜇𝑡+1𝜎𝑡+1
𝑥𝑡−1
𝑧𝑡−1
ℎ𝑡−1
𝜇𝑡−1𝜎𝑡−1
𝑧𝑡 𝑝(𝑧)
ℎ𝑡
𝑥𝑡
𝑝𝜑 𝑥 𝑧
![Page 20: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/20.jpg)
Time Series Embedding
▪ Single historical value not predictive enough
▪ Embedding
– Use lag of ~20 historical observations at every time step
Time steps
Batcht
t +1t +2
![Page 21: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/21.jpg)
Implementation
▪ Implementation in TensorFlow
▪ Running on P100 GPUs for model training
▪ Long time series and large batch sizes require substantial GPU memory
![Page 22: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/22.jpg)
TensorFlow Dynamic RNN
▪ Unrolling rnn with tf.nn.dynamic_rnn
– Simple to use
– Can handle variable sequence length
▪ Not flexible enough for generative networks
![Page 23: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/23.jpg)
TensorFlow Control Structures
▪ Using tf.while_loop
– More to program, need to understand control structures in more detail
– Much more flexible
![Page 24: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/24.jpg)
Implementation
▪ Notations
![Page 25: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/25.jpg)
Implementation
▪ Variable and Weight Setup
Recurrent neuralnetwork definition
![Page 26: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/26.jpg)
Implementation
▪ Allocate TensorArray objects
▪ Fill input TensorArray objects with data
![Page 27: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/27.jpg)
Implementation
▪ While loop body inference part
Update inferencernn state
![Page 28: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/28.jpg)
Implementation
▪ While loop body inference part
Update generatorrnn state
![Page 29: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/29.jpg)
Implementation
▪ Call while loop
▪ Stacking TensorArray objects
![Page 30: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/30.jpg)
Implementation
▪ Loss Calculation
![Page 31: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/31.jpg)
FX Market
▪ FX market is largest and most liquid market in the world
▪ Decentralized over the counter market
– Not necessary to go through a centralized exchange
– No single price for a currency at a given point in time
▪ Fierce competition between market participants
▪ 24 hours, 5 ½ days per week
– As one major forex market closes, another one opens
![Page 32: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/32.jpg)
FX Data
▪ Collect tick data from major liquidity provider e.g. LMAX
▪ Aggregation to OHLC bars (1s, 10s, …)
▪ Focus on US trading session
8am – 5pm EST
3am – 12am EST
5pm – 2am EST (Sidney)
London sessionUS session Asian session
7pm – 4am EST (Tokyo)
5 4 3 2 1 12 11 10 9 8 7 6 5 4 3 2 1 12 11 10 9 8 7 6
![Page 33: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/33.jpg)
EURUSD 2016
![Page 34: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/34.jpg)
Single Day
![Page 35: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/35.jpg)
One Hour
![Page 36: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/36.jpg)
10 Min Sampled at 1s
5 pips
1/10 pips = 1 deci-pip
At high frequency FX prices fluctuate in rangeof deci-pips
Larger jumps in the order ofmultiple pipsand more
![Page 37: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/37.jpg)
Setup
▪ Normalize data with std deviation ො𝜎 over training interval
▪ 260 trading days in 2016, one model per day
▪ 60 dim embedding, 2 dim latent space
ො𝜎Training
Out of sample test
![Page 38: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/38.jpg)
Results
Training
![Page 39: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/39.jpg)
Out of Sample
![Page 40: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/40.jpg)
Volatility of Prediction
![Page 41: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/41.jpg)
Latent Variables
![Page 42: Dive Deeper in Finance · 2017. 5. 18. · Today Generative models for financial time series –Sequential latent Gaussian Variational Autoencoder Implementation in TensorFlow –Recurrent](https://reader035.vdocument.in/reader035/viewer/2022071109/5fe38f693067e045aa0aaa34/html5/thumbnails/42.jpg)
Pricing in E-Commerce
▪ Attend our talk on our latest work on AI and GPU accelerated genetic algorithms with Jet.com