Danilo Mandic
Professor of Machine Intelligence
Designs statistical learning algorithms including tensor and graph learning and interpretable deep networks, applied to hearables and finance.
Danilo Mandic's lab at Imperial works on learning theory & optimization, trustworthy & safe ML and interpretability, evaluation & safety of LMs. 83 papers since 2023, 14 at top venues, mostly ICASSP and IJCNN. Applications close .
At a glance
- papers, 2023–2026
- 22/ year
- at T1 venues
- 17%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom DBLP, 2023–2026
- 13 papers2026 · T1 2 · collecting
- 18 papers2025 · T1 4
- 23 papers2024 · T1 2
- 29 papers2023 · T1 6
T1 venuesOtherVenues: ICASSP, IJCNN, Neural Networks, ICML, Neural Computation
Research focuspaper titles that use each topic's words
- Learning theory & optimization18 papers
- hearables3 papers
- Interpretability, evaluation & safety of LMs2 papers
- Trustworthy & safe ML0 papers
- deep networks0 papers
Paper list
From DBLP.
202613 papers
- Neural NetworksA novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks (opens in a new tab)
- IEEE Signal Process. Mag.Artificial Intelligence for Education: A Signal Processing Perspective: Part II: From Human-AI Cocreativity to Educational Equity [From the Guest Editors] (opens in a new tab)
- AAAI (T1)Coarse-to-Fine Open-Set Graph Node Classification with Large Language Models (opens in a new tab)
- IEEE Signal Process. Mag.Demystifying CNNs: The Role of Domain Knowledge in Enhancing Interpretability and Efficiency [Lecture Notes] (opens in a new tab)
- IEEE Signal Process. Mag.Demystifying Data Augmentation for Multivariate and Nonstationary Time Series: A Fourier transform approach [Lecture Notes] (opens in a new tab)
- IEEE AccessEnhancement of RGB-Based Human Action Recognition via Event-Inspired Multitask Learning (opens in a new tab)
- IEEE Signal Process. Mag.Fair and Skill-Diverse Student Group Formation: A graph-theoretic approach [Special Issue on Artificial Intelligence for Education: A Signal Processing Perspective] (opens in a new tab)
- Digit. Signal Process.Geodesic least mean square algorithm for 3D tracking (opens in a new tab)
- IEEE Trans. Signal Process.QLMS-QHM: Accelerated Quaternion LMS Algorithm With Quasi-Hyperbolic Momentum (opens in a new tab)
- Neural NetworksQuaternion-based CNN for heart rate prediction from PPG (opens in a new tab)
- Mach. Intell. Res.Surviving Resource-constraint Compression: Capability Retention under Tensor-train Decomposition for Sub-billion Parameter Language Models (opens in a new tab)
- ACL (T1)TeRA: Vector-based Random Tensor Network for High-Rank Adaptation of Large Language Models (opens in a new tab)
202518 papers
- Signal Process.A class of widely linear quaternion blind equalisation algorithms (opens in a new tab)
- DSPA Quantum of Learning: Using Quaternion Algebra to Model Learning on Quantum Devices (opens in a new tab)
- IEEE Signal Process. Mag.Artificial Intelligience for Education: A Signal Processing Perspective: Part I: From Active Learning to Mitigating Gender Bias [From The Guest Editors] (opens in a new tab)
- IJCNNAugmentation of EEG and ECG Time Series for Machine Learning Applications: Integrating Changepoint Detection into the iAAFT Surrogates (opens in a new tab)
- IEEE AccessComparison of Linear and Nonlinear Methods for Decoding Selective Attention to Speech From Ear-EEG Recordings (opens in a new tab)
- IEEE Trans. Signal Process.Cramér-Rao Lower Bounds for Unconstrained and Constrained Quaternion Parameters (opens in a new tab)
- IEEE Signal Process. Mag.Demystifying Encoder-Decoder Neural Networks: Correncoder for Regression via Latent Spaces [Lecture Notes] (opens in a new tab)
- ICAIFFinDPO: Financial Sentiment Analysis for Algorithmic Trading through Preference Optimization of LLMs (opens in a new tab)
- ICASSP (T1)Kernel-Based Anomaly Detection Using Generalized Hyperbolic Processes (opens in a new tab)
- NeurocomputingOnline censoring-based learning algorithms for fully complex-valued neural networks (opens in a new tab)
- IEEE Trans. Signal Inf. Process. over NetworksOnline Graph Topology Learning via Time-Vertex Adaptive Filters: From Theory to Cardiac Fibrillation (opens in a new tab)
- Signal Process.Optimizing beamforming in quaternion signal processing using projected gradient descent algorithm (opens in a new tab)
202423 papers
- Optim. Lett.Almost sure convergence of stochastic composite objective mirror descent for non-convex non-smooth optimization (opens in a new tab)
- npj Digit. MedicineArtificial intelligence-enhanced electrocardiography derived body mass index as a predictor of future cardiometabolic disease (opens in a new tab)
- IEEE Signal Process. Mag.Augmented Statistics of Quaternion Random Variables: A lynchpin of quaternion learning machines [Hypercomplex Signal and Image Processing] (opens in a new tab)
- IEEE Trans. Neural Networks Learn. Syst.Comprehensive Graph Gradual Pruning for Sparse Training in Graph Neural Networks (opens in a new tab)
- IEEE Signal Process. Mag.Demystifying the Hypercomplex: Inductive biases in hypercomplex deep learning [Hypercomplex Signal and Image Processing] (opens in a new tab)
- IEEE Trans. Circuits Syst. II Express BriefsDeployment of Artificial Intelligence Models on Edge Devices: A Tutorial Brief (opens in a new tab)
- ICASSP WorkshopDetecting Gamma-Band Responses to the Speech Envelope for the ICASSP 2024 Auditory EEG Decoding Signal Processing Grand Challenge (opens in a new tab)
- SensorsExploring Convolutional Neural Network Architectures for EEG Feature Extraction (opens in a new tab)
- Neural NetworksFinding core labels for maximizing generalization of graph neural networks (opens in a new tab)
- ICAIFFinLlama: LLM-Based Financial Sentiment Analysis for Algorithmic Trading (opens in a new tab)
- IJCNNGPINN: Physics-Informed Neural Network with Graph Embedding (opens in a new tab)
- ICML (T1)Graph-based Time Series Clustering for End-to-End Hierarchical Forecasting (opens in a new tab)
202329 papers
- Neural NetworksA class of doubly stochastic shift operators for random graph signals and their boundedness (opens in a new tab)
- IEEE Signal Process. Lett.A Class of Online Censoring Based Quaternion-Valued Least Mean Square Algorithms (opens in a new tab)
- DSPA comparative study on ML-based approaches for Main Entity Detection in Financial Reports (opens in a new tab)
- SensorsAn In-Ear PPG-Based Blood Glucose Monitor: A Proof-of-Concept Study (opens in a new tab)
- ICML (T1)AudioLDM: Text-to-Audio Generation with Latent Diffusion Models (opens in a new tab)
- ICASSP (T1)ClassA Entropy for the Analysis of Structural Complexity of Physiological Signals (opens in a new tab)
- IEEE Trans. Signal Process.Convex Quaternion Optimization for Signal Processing: Theory and Applications (opens in a new tab)
- IEEE Trans. Syst. Man Cybern. Syst.Convolutional Neural Networks Demystified: A Matched Filtering Perspective-Based Tutorial (opens in a new tab)
- SensorsDesign and Implementation of an Atrial Fibrillation Detection Algorithm on the ARM Cortex-M4 Microcontroller (opens in a new tab)
- EMBCFeasibility of Transfer Learning from Finger PPG to In-Ear PPG (opens in a new tab)
- IEEE Control. Syst. Lett.Filtering Structures for α-Stable Systems (opens in a new tab)
- DSPFinancial News Classification Model for NLP-based Bond Portfolio Construction (opens in a new tab)
Where alumni went
Alumni are not listed on the lab homepage.
Apply to Imperial2027 (Apr/Oct)
Computing PhD
- Round
- Round 2 (1 Dec 2026)
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- none
Confirmed on the program page.
Similar labsnearest by research summary
- USCMU
Tuomas Sandholm
- Learning theory & optimization
- USUW
Radha Poovendran
- Trustworthy & safe ML
- CNTsinghua
Bo Zhang
- Learning theory & optimization
- USUSC
Bart Andrew Kosko
0 T1 - CHETH Zurich
Martin Vechev
- Trustworthy & safe ML
- USStanford
Tengyu Ma
- Learning theory & optimization
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The member list could not be found on the lab site.
- The alumni list could not be found on the lab site.
- Collected
- Verified
- Papers
- DBLP
- Members
- Not collected
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
Imperialcloses