Pith. sign in

REVIEW 5 major objections 6 minor 52 references

Comyco: Quality-Aware Adaptive Video Streaming via Imitation Learning

T0 review · 5 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Imitation training from an instant lookahead solver makes video bitrate policy learning 1700x more sample-efficient and improves perceived quality.

desk verdict Comyco's imitation-learning approach to ABR is promising and well evaluated, but its expert solver's use of future throughput is a caveat reviewers should probe. read the letter →

arxiv 1908.02270 v2 pith:XPE6LBCT submitted 2019-08-06 cs.MM cs.AIcs.NI

Tianchi Huang , Chao Zhou , Rui-Xiao Zhang , Chenglei Wu , Xin Yao , Lifeng Sun This is my paper · ORCID
classification cs.MMcs.AIcs.NI
keywords adaptivebitratestreamingimitationlearningqualityofexperienceVMAFsampleefficiencyreinforcementDASHneuralnetwork
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Comyco is an adaptive-bitrate (ABR) streaming algorithm trained by imitation learning rather than by slow reinforcement-learning exploration. The paper claims that a finite-horizon 'instant solver' can compute near-optimal bitrate choices for any playback state, and that training a neural network to copy those expert actions cuts the required samples by about 1700x and training time by about 16x. It also replaces bitrate with perceptual video quality (VMAF) in the reward and network inputs, reporting 7.5%-16.79% higher average QoE than prior schemes and 7.37% higher average video quality under equal rebuffering time. If true, this makes learning-based bitrate control practical to train and shifts the objective from raw bitrate to what viewers actually perceive.

What carries the argument

The load-bearing machinery is the instant solver coupled to a virtual player. The virtual player simulates the client by advancing virtual time as $t_{k+1}=t_k+d_k(R_k)/C_k+\delta t_k$ and updating buffer occupancy as $B_{k+1}=[(B_k-d_k(R_k)/C_k)^+ + L-\delta t_k]^+$; over a horizon of $N=8$ chunks, the instant solver maximizes the QoE objective over the allowable bitrates and returns the first chunk's choice as the expert action. That expert label, combined with Comyco's loss $L_{\mathrm{comyco}}=-\sum \hat{A}\log\pi(s,a;\theta)-\alpha H(\pi(s;\theta))$, turns every rollout sample into a supervised update toward a near-optimal action, which is why training needs so few samples.

What would settle it

Compare the instant solver's chosen bitrates against the true offline-optimal choices computed with complete trace knowledge; if they disagree on a substantial fraction of decisions, or if real deployments where measured download times deviate from $d_k(R_k)/C_k$ fail to show the reported QoE gains, the imitation premise collapses.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes a method, not just a result: an ABR policy can be trained by imitation instead of by reinforcement learning from scratch. The key move is the instant solver, a finite-horizon ($N=8$) optimizer that, given the current playback state and known future throughput from an offline trace, computes a near-optimal action for the next chunk using the virtual player's buffer model. Comyco's agent rolls out its current policy, asks the solver for the expert action at each visited state, stores the pairs in a replay buffer, and updates the network with a cross-entropy loss against the expert action plus an entropy bonus. The paper claims this avoids the sample inefficiency of policy-gradient RL and the compounding errors of pure supervised cloning. It further claims that replacing bitrate with VMAF in both the QoE metric and the network input makes the learned policy prefer perceptually better chunks, yielding the reported QoE and quality gains.

Load-bearing premise

The expert actions are trustworthy: the virtual player's model of download time and buffer occupancy over an N=8 horizon must be close enough to real playback that copying its choices improves real-world QoE.

Editorial extensions

If this is right

  • If correct, learning-based ABR can be trained with roughly 1700x fewer environment samples and 16x less wall-clock time, making policy retraining feasible as network traces and video libraries change.
  • Bitrate control can be driven by perceptual quality rather than bitrate alone: Comyco reports 7.37% higher average VMAF at the same rebuffering time than the strongest RL baseline.
  • The same policy class no longer needs reward-shaping or long exploration, because expert labels from the instant solver define the learning direction.
  • Quality-aware QoE models built on VMAF correlate more strongly with subjective scores than linear bitrate-based models, supporting their use in ABR evaluation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The imitation-from-emulator recipe generalizes: any control task with a faithful offline simulator and an instant solver could get the same sample-efficiency boost, so congestion control or cache scheduling are natural next targets.
  • The N=8 horizon is a cost/performance compromise, and the paper's ablation shows performance still rises at N=9; adapting the horizon to network volatility is a natural extension.
  • Because the QoE coefficients are fit to a single subjective dataset, transferring Comyco to new devices or content types may require recalibration; the paper's own weaker result on music videos hints that content-dependent behavior is not fully captured.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. Comyco is a learning-based adaptive bitrate (ABR) algorithm for DASH streaming. It trains a neural network policy by imitation learning from an "instant solver" that performs an N-step lookahead over a virtual player to produce near-optimal bitrate decisions. The authors incorporate perceptual video quality into the system by using VMAF as the quality measure, fitting a QoE model to the external Waterloo SQoE-III dataset, and constructing a new video quality dataset. In trace-driven evaluation, Comyco is reported to require 1700x fewer training samples and 16x less training time than a retrained Pensieve, to improve average QoE by 7.5%-16.79% over baselines, and to achieve higher quality under the same rebuffering time. The paper also reports real-world deployment results, ablation studies on the lookahead horizon and entropy coefficient, and computational overhead.

Significance. If the reported gains hold, Comyco would be a valuable step toward making learned ABR policies practical: the imitation-learning formulation avoids the exploration phase of deep RL and makes sample-efficient training plausible, and the quality-aware design is a genuine advance over bitrate-only QoE objectives. Importantly, the QoE objective is anchored to an external subjective dataset (SQoE-III) and an external perceptual metric (VMAF), which mitigates the circularity concern that would arise if both training and evaluation used an invented metric. The authors provide a code repository and evaluate on public traces, which supports reproducibility. The main risk is the non-causal expert signal, which means the policy is trained to imitate decisions based on future network information; this needs to be addressed with causal ablations and more rigorous statistics before the performance claims can be considered fully supported.

major comments (5)
  1. [§4.2 / §4.1] The expert actions produced by the instant solver are not functions of the policy's state. Eq. (2) uses future throughput C_t over the lookahead horizon to pick R_k, whereas the state S_k in §4.1 contains only past throughputs (the C_k vector), video content, and playback features. The paper neither quantifies the resulting label variability nor compares against a causal expert (e.g., one using harmonic-mean throughput prediction as in Robust MPC). Because this expert signal is the training target, the reported 1700x sample-efficiency and QoE improvements may not transfer to deployment; please add a causal-expert ablation and report the variance of expert actions for identical states.
  2. [§6.2 / Fig. 8] The sample-efficiency claims are not backed by a measurement protocol. Figure 8 reports "Epochs" and "Training Time" but does not state the convergence threshold, the number of independent runs, or error bars; the "1700x improvement" and "16x improvement" numbers are therefore not reproducible from the text. Please specify the target performance level used for the comparison and report the distribution of results across seeds.
  3. [§6.2] The Pensieve baseline may not be the published Pensieve algorithm. The text says "we retrain Pensieve via our datasets (§6.1), NN architectures (§4.1) and QoE metrics (§5.1)"; if the architecture is Comyco's, this is a hybrid baseline rather than the Pensieve of Mao et al., and the claim of surpassing Pensieve by 7.37% would not be a comparison with the state-of-the-art system as published. Please compare against the original Pensieve implementation or clearly label and justify the reimplementation.
  4. [§6.5] The real-world deployment results are reported as average QoE without the number of sessions per network condition or confidence intervals. Figure 10 shows only point estimates, and the International link has high throughput variance (sigma = 266.9 KB/s), so the stated improvements of 4.57%-9.93% over Pensieve and 6.43%-9.46% over RobustMPC are not statistically supported. Please report per-session results, error bars, and the number of sessions for each network condition.
  5. [§6.1] The virtual player is described as "faithful" but no validation of its download-time and buffer dynamics against a real client is provided. Since the trace-driven QoE comparisons in §6.2-§6.4 are computed in the same simulator used to train the policy, the fidelity of this player is load-bearing for the relative QoE claims. Please validate the virtual player's predictions (e.g., download time, buffer evolution) against actual playback traces, or explicitly discuss the limitations.
minor comments (6)
  1. [§4.1 / §4.2] The symbol C_k is used both for the vector of past network features in §4.1 and for the throughput at chunk k in Eq. (2), which is confusing; please use distinct notations.
  2. [§4.3] The derivation in Eq. (3) is not a correct statement of the Bellman optimality equation: maximizing V_pi(s) is not equivalent to maximizing q*(s,a) over a. The loss in Eq. (4) stands on its own as behavioral cloning with entropy regularization, but the motivation as written should be corrected.
  3. [§4.7] The learning rate is written as alpha = 10^-4 in §4.7, while alpha is already used as the entropy coefficient in Eq. (4); please use different symbols for these two hyperparameters.
  4. [Fig. 8] The axes of Figure 8 appear to lack labels and units; please add them so the convergence curves are interpretable.
  5. [§5.2] The dataset description reports 394,551 chunks and 1,578,204 quality assessments, which implies four assessments per chunk, but only VMAF, VMAF-4K, and VMAF-phone are listed; please clarify the fourth assessment.
  6. [§7] The related work section cites reference [50] for Indigo, but reference [50] appears to be the Pantheon paper; please check the citation.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the core imitation-learning claim is grounded in external QoE data and independent baselines; self-citations are not load-bearing.

full rationale

Comyco's expert trajectories come from an instant solver that maximizes the QoE_v objective over a finite lookahead (Eq. 2), and the policy is trained to imitate those actions (Eq. 4). Using the same objective for training and evaluation is the standard formulation for imitation learning and RL, and it does not by itself make the comparison circular: the policy is not constructed from the evaluation score, and the reported gains are measured against independently implemented baselines (Pensieve, Robust MPC, BOLA, RB) under a shared trace-driven player. The QoE_v metric is also not an input masquerading as a prediction: its parameters are fit by linear regression on the external Waterloo Streaming SQoE-III database with an 80/20 split and 1,000 runs, and q(Rn) is set to VMAF, an external perceptual quality metric. Table 1 reports held-out SRCC against existing QoE models. The main caveat is that the instant solver uses future throughput Ct inside Eq. 2, which is not present in the policy state Sk described in Section 4.1; this creates a potential causal mismatch and a generalization risk, but it is not a circular reduction because the expert labels are not defined in terms of the learned policy. The two self-citations ([18] Tiyuntsong and [19] QARC) appear in related work and as inspiration for multi-agent parallel training in Section 4.6; neither is invoked to justify the central prediction or to forbid alternative solutions, so they are not load-bearing. Overall, the derivation chain is self-contained against external benchmarks, and no step reduces to its own input by construction.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on three fitted or hand-picked parameters (QoE weights, entropy weight, lookahead horizon) and on the fidelity of the virtual-player/QoE model. No new physical entities are introduced.

free parameters (3)
  • QoE metric coefficients (alpha, beta, gamma, delta) = 0.8469, 28.7959, 0.2979, 1.0610
    Fitted via linear regression on the Waterloo SQoE-III dataset to maximize correlation with MOS; these weights define the QoE objective used by the solver, training, and evaluation (Section 5.1).
  • Entropy coefficient alpha (Equation 4) = 0.001
    Selected via ablation study (Section 6.4, Table 3) to balance exploration and exploitation; directly controls the training loss.
  • Instant solver lookahead horizon N = 8
    Chosen as a performance/computation tradeoff in the ablation (Section 6.4, Table 2); the expert policy quality depends on N.
assumptions (4)
  • domain assumption The virtual player model in Eq. 2 (download time dk(Rk)/Ck and buffer evolution with max buffer Bmax) faithfully reproduces real ABR client behavior.
    Used by the instant solver to compute expert actions and by the evaluation; fidelity is asserted by reference to Pensieve/Oboe/Sabre simulators, not validated against real-world logs in the paper.
  • domain assumption The linear QoE model QoEv with VMAF as q(Rn) tracks real user experience.
    Parameters are fit to SQoE-III, but no test on held-out streaming sessions with MOS is performed; all policy comparisons use this same metric.
  • domain assumption Imitation learning with cross-entropy plus an entropy bonus converges to a policy close to the expert.
    Assumed from prior imitation-learning results; no convergence guarantee is given (Section 4.3).
  • domain assumption VMAF is a valid perceptual video quality metric for the encoding ladder and content types used.
    Borrowed from prior work; used as the quality signal and in the QoE model (Section 5.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Comyco: Quality-Aware Adaptive Video Streaming via Imitation Learning." pith.science (2026). https://pith.science/paper/XPE6LBCT

@misc{pith2026190802270,
  author       = {Pith},
  title        = {Pith review of: Comyco: Quality-Aware Adaptive Video Streaming via Imitation Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XPE6LBCT}},
  note         = {Machine review of arXiv:1908.02270}
}
read the original abstract

Learning-based Adaptive Bit Rate~(ABR) method, aiming to learn outstanding strategies without any presumptions, has become one of the research hotspots for adaptive streaming. However, it typically suffers from several issues, i.e., low sample efficiency and lack of awareness of the video quality information. In this paper, we propose Comyco, a video quality-aware ABR approach that enormously improves the learning-based methods by tackling the above issues. Comyco trains the policy via imitating expert trajectories given by the instant solver, which can not only avoid redundant exploration but also make better use of the collected samples. Meanwhile, Comyco attempts to pick the chunk with higher perceptual video qualities rather than video bitrates. To achieve this, we construct Comyco's neural network architecture, video datasets and QoE metrics with video quality features. Using trace-driven and real-world experiments, we demonstrate significant improvements of Comyco's sample efficiency in comparison to prior work, with 1700x improvements in terms of the number of samples required and 16x improvements on training time required. Moreover, results illustrate that Comyco outperforms previously proposed methods, with the improvements on average QoE of 7.5% - 16.79%. Especially, Comyco also surpasses state-of-the-art approach Pensieve by 7.37% on average video quality under the same rebuffering time.

Figures

Figures reproduced from arXiv: 1908.02270 by the authors.

Figure 1
Figure 1. We evaluate quality-aware ABR algorithm and [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Comyco’s Basic System Work-flow Overview. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Comyco’s NN architecture Overview. which stands for the perceptual video quality metrics for each bitrate of the next chunk. ▷ Video playback features. The last essential feature for describ￾ing the ABR’s state is the current video playback status. The status is represented as Fk = {vk−1 , Bk ,Dk ,mk }, where vk−1 is the perceptual video quality metric for the past video chunk selected, Bk ,Dk are vectors which stan… view at source ↗
Figures from the paper (4 more)
Figure 6
Figure 6. Figure 6: Correlation comparison of video presentation qual [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 7
Figure 7. Figure 7: Comparing Comyco with existing ABR approaches under the HSDPA and FCC network traces. Results are illustrated [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 9
Figure 9. Figure 9: Comparing Comyco with existing ABR approaches [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]
Figure 10
Figure 10. Figure 10: Comparing Comyco with Pensieve and Ro￾bustMPC under the real-world network conditions. We take QoE = 60 as baselines. experience strategy in [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 43 canonical work pages

  1. [1]

    DASH Industry Forum | Catalyzing the adoption of MPEG-DASH

    2019. DASH Industry Forum | Catalyzing the adoption of MPEG-DASH. (2019). https://dashif.org/

    work page 2019

  2. [2]

    HTTP Live Streaming

    2019. HTTP Live Streaming. https://developer.apple.com/streaming/. (2019)

    work page 2019

  3. [3]

    Anne Aaron, Zhi Li, Megha Manohara, Joe Yuchieh Lin, Eddy Chi-Hao Wu, and C-C Jay Kuo. 2015. Challenges in cloud based ingest and encoding for high quality streaming media. In 2015 IEEE International Conference on Image Processing (ICIP) . IEEE, 1732–1736

    work page 2015

  4. [4]

    Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et al

  5. [5]

    Tasnim Abar, Asma Ben Letaifa, and Sadok El Asmi. 2017. Machine learning based QoE prediction in SDN networks. In 2017 13th International Wireless Com- munications and Mobile Computing Conference (IWCMC) . IEEE, 1395–1400

    work page 2017

  6. [6]

    Zahaib Akhtar and et al. 2018. Oboe: auto-tuning video ABR algorithms to network conditions. In SIGCOMM 2018. ACM, 44–58

    work page 2018

  7. [7]

    David M Beazley et al. 1996. SWIG: An Easy to Use Tool for Integrating Scripting Languages with C and C++.. In Tcl/Tk Workshop. 43

    work page 1996

  8. [8]

    Jacob Benesty, Jingdong Chen, Yiteng Huang, and Israel Cohen. 2009. Pearson correlation coefficient. In Noise reduction in speech processing . Springer, 1–4

    work page 2009

Show all 52 references
  1. [9]

    Abdelhak Bentaleb, Ali C Begen, and Roger Zimmermann. 2016. SDNDASH: Improving QoE of HTTP adaptive streaming using software defined networking. In Proceedings of ACM MultiMedia 2016 . ACM, 1296–1305

    2016

  2. [10]

    Abdelhak Bentaleb, Bayan Taani, Ali C Begen, Christian Timmerer, and Roger Zimmermann. 2018. A Survey on Bitrate Adaptation Schemes for Streaming Media over HTTP. IEEE Communications Surveys & Tutorials (2018)

    2018

  3. [11]

    Cisco. 2017. Cisco Visual Networking Index: Forecast and Method- ology, 2016-2021. (2017). https://www.cisco.com/c/dam/en/us/ solutions/collateral/service-provider/visual-networking-index-vni/ complete-white-paper-c11-481360.pdf

    2017

  4. [12]

    Zhengfang Duanmu, Abdul Rehman, and Zhou Wang. 2018. A Quality-of- Experience Database for Adaptive Video Streaming. IEEE Transactions on Broad- casting 64, 2 (June 2018), 474–487

    2018

  5. [13]

    Zhengfang Duanmu, Kai Zeng, Kede Ma, Abdul Rehman, and Zhou Wang. 2017. A quality-of-experience index for streaming video. IEEE Journal of Selected Topics in Signal Processing 11, 1 (2017), 154–166

    2017

  6. [14]

    FFmpeg. [n. d.]. FFmpeg. ([n. d.]). http://ffmpeg.org/

  7. [15]

    Gadaleta, F

    M. Gadaleta, F. Chiariotti, M. Rossi, and A. Zanella. 2017. D-DASH: A Deep Q- Learning Framework for DASH Video Streaming. IEEE Transactions on Cognitive Communications and Networking 3, 4 (Dec 2017), 703–718. https://doi.org/10. 1109/TCCN.2017.2755007

    2017

  8. [16]

    GPAC. [n. d.]. MP4BOX. ([n. d.]). https://gpac.wp.imt.fr/mp4box/

  9. [17]

    Alain Hore and Djemel Ziou. 2010. Image Quality Metrics: PSNR vs. SSIM. (2010), 2366–2369

    2010

  10. [18]

    Tianchi Huang, Xin Yao, Chenglei Wu, Rui-Xiao Zhang, and Lifeng Sun. 2018. Tiyuntsong: A Self-Play Reinforcement Learning Approach for ABR Video Streaming. arXiv preprint arXiv:1811.06166 (2018)

    2018 arXiv

  11. [19]

    Tianchi Huang, Rui-Xiao Zhang, Chao Zhou, and Lifeng Sun. 2018. QARC: Video Quality Aware Rate Control for Real-Time Video Streaming based on Deep Reinforcement Learning. In 2018 ACM Multimedia Conference on Multimedia Conference. ACM, 1208–1216

    2018

  12. [20]

    Berglund, and Zhi Li

    Te-Yuan Huang, Chaitanya Ekanadham, Andrew J. Berglund, and Zhi Li. 2019. Hindsight: Evaluate Video Bitrate Adaptation at Scale. In Proceedings of the 10th ACM Multimedia Systems Conference (MMSys ’19) . ACM, New York, NY, USA, 86–97. https://doi.org/10.1145/3304109.3306219

    2019

  13. [21]

    Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, and Mark Watson. 2015. A buffer-based approach to rate adaptation: Evidence from a large video streaming service. ACM SIGCOMM Computer Communication Review 44, 4 (2015), 187–198

    2015

  14. [22]

    Ahmed Hussein, Mohamed Medhat Gaber, Eyad Elyan, and Chrisina Jayne. 2017. Imitation learning: A survey of learning methods. ACM Computing Surveys (CSUR) 50, 2 (2017), 21

    2017

  15. [23]

    Junchen Jiang, Vyas Sekar, and Hui Zhang. 2014. Improving fairness, efficiency, and stability in http-based adaptive video streaming with festive. TON 22, 1 (2014), 326–340

    2014

  16. [24]

    Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic opti- mization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  17. [25]

    Michael Laskey, Jonathan Lee, Roy Fox, Anca Dragan, and Ken Goldberg. 2017. Dart: Noise injection for robust imitation learning.arXiv preprint arXiv:1703.09327 (2017)

    2017 arXiv

  18. [26]

    Ningning Ma, Xiangyu Zhang, Hai-Tao Zheng, and Jian Sun. 2018. Shufflenet v2: Practical guidelines for efficient cnn architecture design. In Proceedings of the European Conference on Computer Vision (ECCV) . 116–131

    2018

  19. [27]

    Mao. 2017. hongzimao/pensieve. (Jul 2017). https://github.com/hongzimao/ pensieve

    2017

  20. [28]

    Hongzi Mao, Ravi Netravali, and Mohammad Alizadeh. 2017. Neural adaptive video streaming with pensieve. In Proceedings of the 2017 ACM SIGCOMM Con- ference. ACM, 197–210

    2017

  21. [29]

    Hongzi Mao, Shaileshh Bojja Venkatakrishnan, Malte Schwarzkopf, and Mo- hammad Alizadeh. 2019. Variance Reduction for Reinforcement Learning in Input-Driven Environments. international conference on learning representations (2019)

    2019

  22. [30]

    Russell Mendonca, Abhishek Gupta, Rosen Kralev, Pieter Abbeel, Sergey Levine, and Chelsea Finn. 2019. Guided Meta-Policy Search. arXiv preprint arXiv:1904.00956 (2019)

    2019 arXiv

  23. [31]

    Volodymyr Mnih, Adria Puigdomenech Badia, Mehdi Mirza, Alex Graves, Timo- thy Lillicrap, Tim Harley, David Silver, and Koray Kavukcuoglu. 2016. Asynchro- nous methods for deep reinforcement learning. In International Conference on Machine Learning. 1928–1937

    2016

  24. [32]

    Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra, and Martin Riedmiller. 2013. Playing atari with deep reinforcement learning. arXiv preprint arXiv:1312.5602 (2013)

    2013 arXiv

  25. [33]

    Pavlo Molchanov, Stephen Tyree, Tero Karras, Timo Aila, and Jan Kautz. 2016. Pruning convolutional neural networks for resource efficient inference. arXiv preprint arXiv:1611.06440 (2016)

    2016 arXiv

  26. [34]

    Ravi Netravali, Anirudh Sivaraman, Somak Das, Ameesh Goyal, Keith Winstein, James Mickens, and Hari Balakrishnan. 2015. Mahimahi: accurate record-and- replay for HTTP. (2015), 417–429

    2015

  27. [35]

    Takayuki Osa, Joni Pajarinen, Gerhard Neumann, J Andrew Bagnell, Pieter Abbeel, Jan Peters, et al. 2018. An algorithmic perspective on imitation learning. Founda- tions and Trends® in Robotics 7, 1-2 (2018), 1–179

    2018

  28. [36]

    Pablo Gil Pereira, Andreas Schmidt, and Thorsten Herfet. 2018. Cross-Layer Effects on Training Neural Algorithms for Video Streaming. In Proceedings of the 28th ACM SIGMM Workshop on Network and Operating Systems Support for Digital Audio and Video. ACM, 43–48

    2018

  29. [37]

    Yanyuan Qin, Shuai Hao, Krishna R Pattipati, Feng Qian, Subhabrata Sen, Bing Wang, and Chaoqun Yue. 2018. ABR streaming of VBR-encoded videos: char- acterization, challenges, and solutions. In Proceedings of CoNeXT 2018 . ACM, 366–378

    2018

  30. [38]

    Reza Rassool. 2017. VMAF reproducibility: Validating a perceptual practical video quality metric. In Broadband Multimedia Systems and Broadcasting (BMSB), 2017 IEEE International Symposium on . IEEE, 1–2

    2017

  31. [39]

    Abdul Rehman, Kai Zeng, and Zhou Wang. 2015. Display device-adapted video quality-of-experience assessment. In Human Vision and Electronic Imaging XX , Vol. 9394. International Society for Optics and Photonics, 939406

    2015

  32. [40]

    Measuring Fixed Broadband Report. 2016. Raw Data Measuring Broadband Amer- ica 2016. https://www.fcc.gov/reports-research/reports/measuring-broadband- america/raw-data-measuring-broadband-america-2016. (2016). [Online; accessed 19-July-2016]

    2016

  33. [41]

    Haakon Riiser, Paul Vigmostad, Carsten Griwodz, and Pål Halvorsen. 2013. Com- mute path bandwidth traces from 3G networks: analysis and applications. In Proceedings of the 4th ACM Multimedia Systems Conference . ACM, 114–118

    2013

  34. [42]

    Stéphane Ross, Geoffrey Gordon, and Drew Bagnell. 2011. A reduction of imi- tation learning and structured prediction to no-regret online learning. In Pro- ceedings of the fourteenth international conference on artificial intelligence and statistics. 627–635

    2011

  35. [43]

    Kevin Spiteri, Ramesh Sitaraman, and Daniel Sparacio. 2018. From theory to prac- tice: improving bitrate adaptation in the DASH reference player. In Proceedings of the 9th MMSys . ACM, 123–137

    2018

  36. [44]

    Kevin Spiteri, Rahul Urgaonkar, and Ramesh K Sitaraman. 2016. BOLA: Near- optimal bitrate adaptation for online videos. In INFOCOM 2016, IEEE. IEEE, 1–9

    2016

  37. [45]

    Richard S Sutton and Andrew G Barto. 2018. Reinforcement learning: An intro- duction. MIT press

    2018

  38. [46]

    F. Tang, B. Mao, Z. M. Fadlullah, N. Kato, O. Akashi, T. Inoue, and K. Mizutani. 2018. On Removing Routing Protocol from Future Wireless Networks: A Real-time Deep Learning Approach for Intelligent Traffic Control. IEEE Wireless Communications 25, 1 (February 2018), 154–160. h...

    2018

  39. [47]

    Yuan Tang. 2016. TF. Learn: TensorFlow’s high-level module for distributed machine learning. arXiv preprint arXiv:1612.04251 (2016)

    2016 arXiv

  40. [48]

    Usc-Nsl. 2018. USC-NSL/Oboe. (Oct 2018). https://github.com/USC-NSL/Oboe

    2018

  41. [49]

    Zhou Wang. 2017. Video QoE: Presentation Quality vs. Playback Smoothness. (Jul 2017). https://www.ssimwave.com/science-of-seeing/ video-quality-of-experience-presentation-quality-vs-playback-smoothness/

    2017

  42. [50]

    Francis Y Yan, Jestin Ma, Greg D Hill, Deepti Raghavan, Riad S Wahby, Philip Levis, and Keith Winstein. 2018. Pantheon: the training ground for Internet congestion-control research. In 2018{USENIX} Annual Technical Conference ({USENIX}{ ATC} 18). 731–743

    2018

  43. [51]

    Xiaoqi Yin, Abhishek Jindal, Vyas Sekar, and Bruno Sinopoli. 2015. A control- theoretic approach for dynamic adaptive video streaming over HTTP. In ACM SIGCOMM Computer Communication Review . ACM, 325–338

    2015

  44. [2016]

    In OSDI, Vol

    TensorFlow: A System for Large-Scale Machine Learning.. In OSDI, Vol. 16. 265–283

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.