👩🏼‍🏫 Awards and Certificates

Wed, 22 Oct 2025 00:00:00 +0000

🏆 Award: 2025 Best European Master’s Thesis in Mathematical Finance.

Recipient of the 2025 Natixis Award for Best Master’s Thesis in Mathematical Finance, awarded by the . | |

Below is the research paper associated with this award:

Equal contribution

, Imad Idboufous (2024). Solving The Dynamic Volatility Fitting Problem: A Deep Reinforcement Learning Approach. In IJCNN.

PDF Slides Poster

Video

Here I share the Full ceremony replay: | |

Dailymotion:

Video file

Videos Videos of some annimation:

Training Skew	Volatility Fitting

Podcast

{{< audio src="ambient-piano.mp3" >}}

Try it out:

Math block:

$$ \textcolor{red}{\text{Action network:}} \quad \text{DPG} = \textcolor{blue}{\text{Deep Policy Gradient}} $$$$ a_t \sim \textcolor{red}{\pi^{D}(s_t,\theta^{\pi})} + \epsilon_t, \quad \text{with} \quad \textcolor{brown}{\epsilon_t \sim \mathcal{N}(0, \sigma_n^2 I_K)}, \quad \text{and} \quad \sigma_n = \max\!\left(\sigma_0\left(1-\frac{n}{N}\right)^4,\sigma_{\min}\right) $$$$ \textcolor{red}{\text{Critic network:}} \quad \textcolor{blue}{\text{Q-Learning and Bellman equation}} $$$$ \begin{cases} R_{t}=\sum_{i=t}^{T}\gamma^{(i-t)} r(s_{i},a_{i}) \\[6pt] Q^{\pi}(s_{t},a_{t})=\mathbb{E}[R_{t}\mid s_{t},a_{t}] \end{cases} \quad \Rightarrow \quad \textcolor{red}{ L(\theta^{Q})= \mathbb{E}\left[ \left( Q^{\pi}(s_{t},a_{t}; \theta^{Q})- \left(r(s_{t},a_{t})+\gamma Q^{\pi}(s_{t+1},a_{t+1};\theta^{Q})\right) \right)^{2} \right] } $$

Latex code

\begin{itemize}
 \item \textcolor{red}{Action network:} DPG= \textcolor{blue}{Deep Policy gradient} % $ J = \mathbb{E}[R_{s_{t}}] ==== $ (\textbf{Proof:} Cf. Report)
\end{itemize}
$$
a_t \sim \textcolor{red}{\pi^{D} (s_t,\theta^{\pi})} + \epsilon_t \quad \text{with} \quad \textcolor{brown}{\epsilon_t \sim \mathcal{N}(0, \sigma_n^2 I_K)} \quad \text{and} \quad \sigma_n = \text{max}(\sigma_0(1-\frac{n}{N} )^{4},\sigma_{\text{min}})
$$
\begin{itemize}
 \item \textcolor{red}{Critic network:} \textcolor{blue}{Q-Learning and Bellman equation.} % $ Q_{\theta^{Q}}(s_{t_{i}},a_{t_{i}}) = - \sum_{k=t_i}^{T} \mathbb{E}_{(s_{k},a_{k})\sim \rho_\pi} [ \gamma^{(k-t_i)} \xi (\vec{\theta}_{t_{k}} )] $
\end{itemize}
$$
\begin{cases} R_{t}=\sum_{i=t}^{T}\gamma^{(i- t)}r(s_{i},a_{i})\\Q^{\pi}(s_{t},a_{t})=\mathbb{E}[R_{t}|s_{t},a_{t}]&
\end{cases} \quad \Rightarrow \textcolor{red}{L(\theta^{Q})=\mathbb{E}[(Q^{\pi}(s_{t},a_{t}; \theta^{Q})-[r(s_{t},a_{t})+\gamma Q^{\pi}(s_{t+1},a_{t+1};\theta^{Q})])^{2}]}
$$

Stochastic Environment |

👩🏼‍🏫 Awards and Certificates

🏆 Award: 2025 Best European Master’s Thesis in Mathematical Finance.

Related Work

Video

Podcast

Math block: