Christos Bouganis
Professor of Intelligent Digital Systems
Designs high-performance, power-efficient FPGA-based digital systems for machine learning and computer vision.
Christos Bouganis's lab at Imperial works on accelerators, chips & architecture and efficient training & inference. 27 papers since 2023, 6 at top venues, mostly ICLR and IJCV. 7 PhD students. 27 alumni, most often to nCompass Technologies and Samsung. Applications close .
At a glance
- papers, 2023–2026
- 7/ year
- at T1 venues
- 22%
- PhD students
- 7
- MS students
- 0
- alumni listed
- 27
Papers by yearfrom DBLP, 2023–2026
- 4 papers2026 · T1 2 · collecting
- 6 papers2025 · T1 2
- 7 papers2024 · T1 1
- 10 papers2023 · T1 1
T1 venuesOtherVenues: ICLR, IJCV, BMVC, ECCV, ICCV
Research focuspaper titles that use each topic's words
- Accelerators, chips & architecture8 papers
- Efficient training & inference6 papers
- high-performance0 papers
- digital0 papers
- designs0 papers
Paper list
From DBLP.
20264 papers
- ECCV (T1)Region-Aware Test-Time Scaling for Compositional Image Generation (opens in a new tab)
- FCCMTowards Small Language Models on FPGAs (opens in a new tab)
- ASAPWhen Timing is Uncertain, Infer: Probabilistic Early FPGA Timing Analysis for Robust Critical Path Optimization (opens in a new tab)
- IJCV (T1)Δ-DiT: Accelerating Diffusion Transformers without Training via Denoising Property Alignment (opens in a new tab)
20256 papers
- DATEA Resource-Aware Residual-Based Gaussian Belief Propagation Accelerator Toolflow (opens in a new tab)
- ICLR (T1)Cached Multi-Lora Composition for Multi-Concept Image Generation (opens in a new tab)
- J. Braz. Comput. Soc.Efficient Number of Functional Units and Loop Pipeline Design Space Exploration for High-Level Synthesis (opens in a new tab)
- ADMAHarmonia: A Swift and Accurate Approximate Data Structure for Real-Time Heavy Flow Detection in High-Speed Networks (opens in a new tab)
- FCCMITERA-LLM: Boosting Sub-8-Bit Large Language Model Inference via Iterative Tensor Decomposition (opens in a new tab)
- ICLR (T1)Towards Understanding Why Label Smoothing Degrades Selective Classification and How to Fix It (opens in a new tab)
20247 papers
- DATEA Framework for Designing Scalable Gaussian Belief Propagation Accelerators for use in SLAM (opens in a new tab)
- IJCV (T1)Augmenting the Softmax with Additional Confidence Scores for Improved Selective Classification with Out-of-Distribution Data (opens in a new tab)
- DATEAuto WS: Automate Weights Streaming in Layer-Wise Pipelined DNN Accelerators (opens in a new tab)
- BMVCBudget-aware Dynamic Spatially Adaptive Inference (opens in a new tab)
- FPLHASS: Hardware-Aware Sparsity Search for Dataflow DNN Accelerator (opens in a new tab)
- WACV WorkshopLatency Driven Spatially Sparse Optimization for Multi-Branch CNNs for Semantic Segmentation (opens in a new tab)
- FCCMSMOF: Streaming Modern CNNs on FPGAs with Smart Off-Chip Eviction (opens in a new tab)
202310 papers
- FCCMATHEENA: A Toolflow for Hardware Early-Exit Network Automation (opens in a new tab)
- ASAPFMM-X3D: FPGA-Based Modeling and Mapping of X3D for Human Action Recognition (opens in a new tab)
- FPLfpgaHART: A Toolflow for Throughput-Oriented Acceleration of 3D CNNs for HAR onto FPGAs (opens in a new tab)
- FCCMHARFLOW3D: A Latency-Oriented 3D-CNN Accelerator Toolflow for HAR on FPGA Devices (opens in a new tab)
- FPLMixed-TD: Efficient Neural Network Accelerator with Layer-Specific Tensor Decomposition (opens in a new tab)
- ComputerMultiple-Deep Neural Network Accelerators for Next-Generation Artificial Intelligence Systems (opens in a new tab)
- FPLPASS: Exploiting Post-Activation Sparsity in Streaming Architectures for CNN Acceleration (opens in a new tab)
- ICFPTSATAY: A Streaming Architecture Toolflow for Accelerating YOLO Models on FPGA Devices (opens in a new tab)
- WACVSVD-NAS: Coupling Low-Rank Approximation and Neural Architecture Search (opens in a new tab)
- ICCV (T1)Window-Based Early-Exit Cascades for Uncertainty Estimation: When Deep Ensembles are More Efficient than Single Models (opens in a new tab)
Where alumni went27 listed, 83% to industry
Industry 5Academia 1
Most often to
- nCompass Technologies
- Samsung
- Apple
Also to
- Hunan Normal University
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
- GBUCL
Adnan Mehonic
- Accelerators, chips & architecture
- CNSJTU
Fangxin Liu
- Accelerators, chips & architecture
- USStanford
Kunle Olukotun
- Accelerators, chips & architecture
- USNYU
Brandon Reagen
- Accelerators, chips & architecture
- CAToronto
Gennady Pekhimenko
- Accelerators, chips & architecture
- USUMD
Bahar Asgari
- Accelerators, chips & architecture
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- DBLP
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Imperialcloses