{
  "schema_version": "0.1.0",
  "record_type": "research-topic",
  "topic_id": "deep-learning-systems",
  "label": "Deep Learning Systems",
  "title": "Deep Learning Systems Research",
  "description": "Deep Learning Systems research papers in the Naser Ezzati-Jivan publication catalog.",
  "introduction": "This topic page groups Naser Ezzati-Jivan research papers related to deep learning systems. Each linked record provides the paper's problem, method, findings, limitations, keywords, and authoritative source links.",
  "aliases": [
    "deep learning systems"
  ],
  "search_terms": [
    "Deep Learning Systems"
  ],
  "related_topics": [],
  "canonical_url": "https://threadslab.org/research-publications/topics/deep-learning-systems.html",
  "paper_count": 4,
  "papers": [
    {
      "paper_id": "deba-adaptive-batch-scheduling",
      "title": "One Size Does Not Fit All: Architecture-Aware Adaptive Batch Scheduling with DEBA",
      "year": 2025,
      "authors": [
        "François Belias",
        "Naser Ezzati-Jivan",
        "Foutse Khomh"
      ],
      "page_url": "https://threadslab.org/research-publications/papers/deba-adaptive-batch-scheduling/",
      "canonical_source_url": "https://arxiv.org/abs/2511.03809",
      "core_contribution": "DEBA is an architecture-aware adaptive batch scheduler that uses training-stability signals to decide when and how to change batch size.",
      "tags": [
        "deep-learning-systems",
        "adaptive-scheduling",
        "performance-optimization",
        "architecture-aware-methods"
      ],
      "keywords": [
        "DEBA",
        "adaptive batch size",
        "gradient variance",
        "training speedup",
        "ResNet",
        "DenseNet",
        "EfficientNet",
        "ViT"
      ]
    },
    {
      "paper_id": "iot-anomaly-classification-cnn-bilstm-two-tiered",
      "title": "A Two-Tiered Framework for Anomaly Classification in IoT Networks Utilizing CNN-BiLSTM Model",
      "year": 2024,
      "authors": [
        "Yue Guan",
        "Morteza Noferesti",
        "Naser Ezzati-Jivan"
      ],
      "page_url": "https://threadslab.org/research-publications/papers/iot-anomaly-classification-cnn-bilstm-two-tiered/",
      "canonical_source_url": "https://doi.org/10.1016/j.simpa.2024.100646",
      "core_contribution": "The paper proposes a two-tier IoT intrusion detector that first separates normal from anomalous flows and then classifies the attack type with a CNN-BiLSTM model.",
      "tags": [
        "iot-security",
        "anomaly-detection",
        "machine-learning",
        "deep-learning-systems"
      ],
      "keywords": [
        "IoT anomaly detection",
        "CNN-BiLSTM",
        "SMOTE",
        "particle swarm optimization",
        "PSO",
        "Software Impacts",
        "classification"
      ]
    },
    {
      "paper_id": "cnn-bilstm-rpl-attacks-iot-smart-grid",
      "title": "CNN-BiLSTM-Based Classification of RPL Attacks in IoT Smart Grid Networks (Industry Track)",
      "year": 2023,
      "authors": [
        "Yue Guan",
        "Morteza Noferesti",
        "Naser Ezzati-Jivan"
      ],
      "page_url": "https://threadslab.org/research-publications/papers/cnn-bilstm-rpl-attacks-iot-smart-grid/",
      "canonical_source_url": "https://doi.org/10.1145/3626562.3626832",
      "core_contribution": "The paper applies a CNN-BiLSTM intrusion classifier to RPL/IoT traffic, combining convolutional feature extraction with bidirectional sequence modeling after imbalance-aware flow preprocessing.",
      "tags": [
        "iot-security",
        "deep-learning-systems",
        "machine-learning",
        "anomaly-detection"
      ],
      "keywords": [
        "RPL attacks",
        "IoT smart grid",
        "CNN-BiLSTM",
        "routing attacks",
        "intrusion detection",
        "Middleware 2023"
      ]
    },
    {
      "paper_id": "iot-anomaly-intrusion-detection-poster-abstract",
      "title": "Deep Learning Driven Anomaly Based Intrusion Detection System for IoT: Poster Abstract",
      "year": 2022,
      "authors": [
        "Yue Guan",
        "Naser Ezzati-Jivan"
      ],
      "page_url": "https://threadslab.org/research-publications/papers/iot-anomaly-intrusion-detection-poster-abstract/",
      "canonical_source_url": "https://doi.org/10.1145/3565386.3565493",
      "core_contribution": "The poster proposes a hybrid IoT intrusion-detection pipeline with binary anomaly detection followed by multiclass attack classification.",
      "tags": [
        "iot-security",
        "anomaly-detection",
        "machine-learning",
        "deep-learning-systems"
      ],
      "keywords": [
        "IoTID20",
        "RNN",
        "SMOTE",
        "PSO",
        "binary classification",
        "multiclass classification",
        "IoT attacks",
        "intrusion detection"
      ]
    }
  ]
}