I work on deep learning, mostly on what happens inside the model.
I take neural networks apart to see what their features encode, then point what I find at EEG signals and robot arms. Most projects start as a question about why something worked and run longer than planned.
- 01Research
- 05
- 02Projects
- 18
- 03Writing
- 09
this is my pfp btw :3
I study what neural networks learn and use it in robot control and EEG decoding.
- 01
Representation Learning
I probe the features a model learns and adapt them to new languages with LoRA-based methods.
- 02
Generative Modeling
I train flow-matching models for speech synthesis and use diffusion models to turn EEG embeddings into images.
- 03
Robotics
I write inverse kinematics for robot arms and compare control mappings for brain-controlled robots.
- 04
Computational Neuroscience
I decode motor intent and seen images from EEG recordings.
- Self-Supervised Learning
- World Models
- Neuroscience
- Neurotechnology
- Neuro-robotics
- Agentic AI
- Language Models
- Quantum Computing
- Nanotechnology
- Nuclear Fusion
Artificial Intelligence
I work on how pretrained models adapt to tasks they were not trained for. For language models I use parameter-efficient fine-tuning: a mixture of LoRA experts for English-to-Indonesian transfer, and LoRA adapters that learn new classes without replaying old data. On the generative side I train flow-matching models for speech. For medical imaging I add Hessian-derived geometric features to a U-Net that segments cell nuclei from a training set of only 30 histopathology images.
- Representation learning
- Diffusion and flow-matching models
- LLM adaptation & PEFT (LoRA, X-LoRA)
- Continual learning
- Medical image segmentation
- PyTorch
- TensorFlow
- Transformers
- Unsloth
- scikit-learn
- WandB
Slide 1 of 5: Control mapping as a design variable: pricing the posterior-to-command step in motor-imagery BCI
Sepuluh Nopember Institute of Technology
AI Engineering Student (Informatics Engineering Dept.)
Relevant coursework: Deep Learning, Machine Learning, Computer Vision, Data Mining, Natural Language Processing, Digital Image Processing, Computational Intelligence, Calculus I & II, Linear Algebra, Probability & Statistics, Discrete Math, Computer Graphics, Object Oriented Programming.
Avalon AI Group
Data Science and Artificial Intelligence Division
Schematics ITS
Expert Staff Data Science Competition
Avalon AI Group
Member
- Explored ML tools and frameworks for data processing, modeling, and deployment.
- Followed recent AI/ML developments through structured community learning sessions.
- Developed robotic programming modules for UAV systems and their integration.
- Collaborated across divisions in assembling and testing a functional UAV unit.